Is SurfMind safe to use?

Yes! SurfMind is designed with security and privacy at its core. SurfMind has been thoroughly reviewed and approved by both the Chrome Web Store and Apple App Store, ensuring it meets strict security and privacy standards.

Do I need an AI API key?

Not necessary. You can either use your own API key from your chosen AI provider or use SurfMind credits instead so you don't have to manage API keys.

How does the AI understand my page?

When you interact with SurfMind, it analyzes the content of the current webpage and includes relevant context in your conversation with the AI. This allows the AI to provide intelligent, context-aware responses.

What browsers are supported?

SurfMind works on all major browsers including Chrome, Firefox, Edge, Brave, Arc, Opera, and Safari. The extension adapts to your browser automatically.

Can I use multiple AI models?

Yes! You can easily switch between different AI models (GPT, Claude, Gemini, etc.) within the extension to find the best one for your specific task.

Does SurfMind support local models or custom AI providers?

Yes. SurfMind fully supports local models and custom endpoints compatible with the OpenAI Completions API.

Models

Browse every AI model available on SurfMind and compare pricing, context window, and speed.

Compare Leaderboard

406 models

DeepSeek V4 Flash Vision Exp
DeepSeek V4 Flash Vision Exp is an experimental multimodal variant of DeepSeek V4 Flash 0731, a sparse MoE model (284B total, 13B active parameters) with a 1M-token context window that now accepts image input alongside text. It matches V4 Flash on agentic reasoning and world knowledge while adding document and chart understanding, visual question answering, and multimodal agent workflows, positioning it as the multimodal entry in DeepSeek's lightweight Flash line.
Aug 21, 20261M context0.1x cost rate
Ox Alpha
Ox Alpha is a stealth-mode reasoning model from an anonymous third-party provider being previewed on OpenRouter, supporting text, image, and video input with a 1M-token context window plus function calling and structured JSON output. It targets coding, sustained agentic work, and long-horizon software engineering, and is offered free during this preview with the provider stating prompts and completions are not used for training.
Aug 20, 20261M contextFree
Hy-MT2-1.8B
Hy-MT2-1.8B is Tencent Hunyuan's smallest open-weight translation model (1.8B dense parameters, Apache 2.0 license), supporting 33 languages as the lightest of the three Hy-MT2 'fast-thinking' sizes alongside the 7B dense and 30B-A3B MoE variants. Despite its size it beats mainstream commercial translation APIs on general and business-domain benchmarks, and AngelSlim quantization down to about 440MB makes it deployable on mobile chipsets for on-device translation.
Aug 20, 20268k context0.1x cost rate
Hy-MT2-30B-A3B
Hy-MT2-30B-A3B is Tencent Hunyuan's largest translation model, a sparse mixture-of-experts architecture (30B total parameters, 3B active, 128 routed plus 1 shared expert) supporting 33 languages, released open-weight (Apache 2.0) alongside the smaller 1.8B and 7B dense Hy-MT2 variants. In fast-thinking mode it outperforms larger open-source general models such as DeepSeek-V4-Pro and Kimi K2.6 on translation benchmarks, making it the pick for high-quality business and domain-specific translation where the dense variants fall short.
Aug 20, 20268k context0.1x cost rate
GLM Latest
This is an alias that always points to Z.ai's latest GLM model.
Aug 19, 20261M context1x cost rate
GLM 5.3
GLM-5.3 is Z.ai's open-weight flagship model (743B parameters, 1M-token context), an incremental upgrade over GLM-5.2 with stronger coding and much better token efficiency per task. Reasoning is always on with low, high, and max effort levels (max by default), and Z.ai positions it as a top open-weight model for complex software engineering and long-horizon agent tasks.
Aug 18, 20261M context1x cost rate
Qwen3.8 27B
Qwen3.8 27B is a smaller, Apache 2.0-licensed open-weight model from Alibaba's Qwen3.8 generation, suited for general-purpose chat, coding, and agentic tasks at moderate cost.
Aug 14, 20261M context0.6x cost rate
Dots3-Note Preview (free)
Dots3-Note Preview is an open-weight mixture-of-experts model from Dots Studio (280B total, 16B active) with a 512K token context window and multimodal input support. It is the lightest model in the dots3 family, suited for coding, long-context document processing, multimodal understanding, and multi-step agentic workflows; this is the free-tier variant.
Aug 14, 2026512k contextFree
Gemini 3.7 Flash
Gemini 3.7 Flash is Google's most intelligent workhorse model yet for coding and agents, an iterative refinement of 3.6 Flash with notably higher accuracy on debugging and issue resolution at roughly half the cost. It is designed for responsive multi-step agentic workflows, coding, and everyday knowledge work.
Aug 13, 20261M context0.4x cost rate
Gemini 3.7 Flash (batch)
Gemini 3.7 Flash is Google's most intelligent workhorse model yet for coding and agents, an iterative refinement of 3.6 Flash with notably higher accuracy on debugging and issue resolution. This is the batch variant for asynchronous, lower-cost processing.
Aug 13, 20261M context0.2x cost rate
Seed 2.1 Turbo
Seed 2.1 Turbo is a multimodal model from ByteDance Seed built for coding and long-horizon agent workflows, including end-to-end software delivery, multi-step task execution, and video/image understanding. It supports planning, debugging, and self-correction with a 262K-token context window.
Aug 12, 2026262k context0.5x cost rate
Qwen3.8 2.4T A95B
Qwen3.8 2.4T-A95B is Alibaba's massive open-weight mixture-of-experts flagship (2.4 trillion total parameters, 95B active), representing the top of the Qwen3.8 generation for frontier-level reasoning, coding, and agentic tasks.
Aug 12, 20261M context1x cost rate
Seed-2.0-Code
Seed 2.0 Code is ByteDance Seed's model optimized specifically for agentic coding, suited to frontend development and multilingual programming tasks. It is designed to work well inside coding-agent tools such as Claude Code, Kilo, and OpenCode.
Aug 12, 2026262k context0.6x cost rate
DeepSeek V4 Pro 0813
DeepSeek V4 Pro (0813) is the August 13, 2026 general-availability checkpoint of DeepSeek V4 Pro, adding DSpark speculative decoding, reasoning-effort levels, and native OpenAI Responses API support. It is DeepSeek's flagship model for coding, reasoning, and agentic tasks in the V4 generation.
Aug 12, 20261M context0.8x cost rate
Grok 4.6
Grok 4.6 is xAI's flagship model built for long-running agentic work and ambitious interactive and visual tasks, with a 500,000-token context window and text/image input. It is tuned for advanced reasoning, coding, and complex agent workflows at the top of xAI's model lineup.
Aug 12, 2026500k context1x cost rate
LFM2.5-2.6B (free)
LFM2.5-2.6B is Liquid AI's on-device agentic model (2.6B parameters, 128K context), pre-trained on ~34T tokens to plan, call tools, and run multi-step tasks entirely on phones, laptops, and edge devices with no data leaving the device. It leads its size class on instruction-following and tool-use benchmarks while running in under 2.5 GB of memory; this is OpenRouter's rate-limited free endpoint.
Aug 11, 202666k contextFree
Nemotron 3.5 Lightning
Nemotron 3.5 Lightning is NVIDIA's open 30B mixture-of-experts model with 3B active parameters and a 1M-token context window, built for fast, accurate execution within long-running agent pipelines. It is designed as a low-latency, low-cost execution layer for always-on agentic systems.
Aug 11, 2026262k context0.1x cost rate
Nemotron 3.5 Lightning (free)
Nemotron 3.5 Lightning is NVIDIA's open 30B mixture-of-experts model with 3B active parameters and a 1M-token context window, built as a low-latency, low-cost execution layer for always-on agentic systems. This is the free-tier variant of the same model.
Aug 11, 20261M contextFree
Sakana Namazu
Sakana Namazu is an OpenAI-compatible model from Sakana AI, built as a fine-tune of Moonshot AI's open-weight Kimi model and tailored for Japanese language and business use cases. It suits Japanese-language chat, business writing, and localized assistant tasks.
Aug 11, 2026262k context0.8x cost rate
Solar Pro 4
Solar Pro 4 is Upstage's next-generation flagship model, an agent-first successor to Solar Pro 3 focused on further reasoning gains at low cost. It targets multi-step tool use and production-scale agentic workloads while retaining Upstage's efficiency-focused, compact-parameter design philosophy.
Aug 10, 2026524k context0.1x cost rate
Muse Glimmer 30B
Muse Glimmer 30B is Meta's open-weight multimodal model, a distilled version of the larger Muse Spark model, combining an LLM with a dedicated perception encoder for text and visual input. It is built for local agentic and coding workloads, and can run entirely offline on a single consumer GPU.
Aug 9, 2026131k context0.2x cost rate
Muse Spark 1.2
Muse Spark 1.2 is an updated release in Meta Superintelligence Labs' Muse family, sharing Muse Spark's focus on multimodal reasoning, coding, and AI-assisted software development with long-context support. It offers incremental improvements over Muse Spark 1.1 for agentic and coding workflows.
Aug 5, 20261M context0.9x cost rate
Qwen3.8 Max
Qwen3.8 Max is Alibaba's proprietary flagship-tier model in the Qwen3.8 generation, intended for top-tier reasoning, coding, and agentic performance alongside the open-weight Qwen3.8 releases.
Aug 3, 20261M context1x cost rate
DeepSeek V4 Flash Latest
This is an alias that always points to DeepSeek's latest V4-Flash model.
Aug 1, 20261.3M context0.1x cost rate
DeepSeek V4 Flash 0731
DeepSeek V4 Flash (0731) is the July 31, 2026 public-beta checkpoint of DeepSeek's V4 Flash model, DeepSeek's fast, cost-efficient V4-generation model tuned for agentic workflows. It offers the same lightweight, high-throughput design as other V4 Flash releases.
Jul 31, 20261.3M context0.1x cost rate
Inkling Small
Inkling-Small is a compact open-weights sibling of Thinking Machines Lab's Inkling model, at 276B total parameters (12B active) versus Inkling's 975B. It reasons natively over text, images, and audio with adjustable thinking effort, achieving performance close to the full Inkling model at a fraction of its size.
Jul 30, 20261M context0.3x cost rate
Qwen3.7 Flash
Qwen3.7 Flash is Alibaba's fast, low-cost proprietary tier of the API-only Qwen3.7 generation, built for high-throughput, latency-sensitive tasks.
Jul 27, 20261M context0.1x cost rate
Claude Opus 5 (Fast)
Fast-mode variant of Claude Opus 5, Anthropic's Claude 5 flagship, offering the same top-tier coding and reasoning capability with reduced latency for time-sensitive agentic workloads.
Jul 24, 20261M context10x cost rate
Claude Opus 5
Claude Opus 5 is Anthropic's flagship model in the Claude 5 generation, representing the top of Anthropic's intelligence tier for coding, agentic workflows, and complex reasoning. It targets the most demanding professional software engineering and long-horizon agent tasks Anthropic offers.
Jul 24, 20261M context5x cost rate
Claude Opus 5 (batch)
Batch-processing variant of Claude Opus 5, Anthropic's Claude 5 generation flagship model. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Jul 24, 20261M context2.5x cost rate
Ling-3.0-flash
Ling 3.0 Flash is InclusionAI's (Ant Group) successor to Ling 2.6 Flash, a hybrid-reasoning Mixture-of-Experts model (~124B total, ~5.1B active) combining Kimi Delta Attention with Multi-Head Latent Attention for efficient long-range memory. It supports both thinking and non-thinking modes with a native 262K context, suited for fast general-purpose coding and agentic tasks.
Jul 23, 2026262k context0.1x cost rate
Laguna S 2.1
Laguna S 2.1 is an open-weight agentic coding model from Poolside, a 118B-parameter mixture-of-experts system with 8B active parameters per token and up to a 1M-token context window. Trained with reinforcement learning inside Poolside's own agent harness, it targets agentic software engineering tasks and is competitive with much larger models on SWE-bench and Terminal-Bench.
Jul 21, 20261M context0.1x cost rate
Laguna S 2.1 (free)
Free-tier variant of Poolside's Laguna S 2.1, an open-weight 118B MoE agentic coding model with a 1M-token context window trained via RL in Poolside's agent harness. Same underlying model as the paid version, suited for agentic software engineering and terminal/coding tasks.
Jul 21, 2026262k contextFree
Gemini 3.6 Flash
Gemini 3.6 Flash is Google's workhorse model delivering a step up in coding, knowledge work, and multimodal performance over 3.5 Flash, with meaningfully improved token efficiency. It targets everyday agentic and coding tasks that need strong quality at Flash-tier cost.
Jul 21, 20261M context0.8x cost rate
Gemini 3.6 Flash (batch)
Gemini 3.6 Flash is Google's workhorse model delivering a step up in coding, knowledge work, and multimodal performance with improved token efficiency. This is the batch variant for asynchronous, lower-cost processing.
Jul 21, 20261M context0.4x cost rate
Gemini 3.5 Flash Lite
Gemini 3.5 Flash-Lite is the most cost-effective model in Google's 3.5 Flash line, built for high-volume, low-latency tasks where price and speed are the priority. It trades peak reasoning depth for efficiency at scale.
Jul 21, 20261M context0.5x cost rate
Gemini 3.5 Flash Lite (batch)
Gemini 3.5 Flash-Lite is the most cost-effective model in Google's 3.5 Flash line, built for high-volume, low-latency tasks where price and speed are the priority. This is the batch variant for asynchronous, lower-cost processing.
Jul 21, 20261M context0.2x cost rate
LongCat 2.0
LongCat-2.0 is Meituan's open-weight 1.6-trillion-parameter Mixture-of-Experts model (~48B active) with a native 1M-token context window, using sparse attention for efficient long-context processing. It targets agentic coding, repository-level code changes, and long-horizon problem solving, with reported performance comparable to Gemini 3.1 Pro on coding benchmarks.
Jul 20, 20261M context0.3x cost rate
Inkling
Inkling is Thinking Machines Lab's first flagship model, a 975B-parameter (41B active) open-weights mixture-of-experts model that reasons natively over text, images, and audio with up to a 1M-token context window. It is designed as a general-purpose frontier model for multimodal reasoning and agentic tasks, released under Apache 2.0.
Jul 17, 20261M context0.8x cost rate
Inkling (batch)
This is the batch-processing variant of Thinking Machines Lab's Inkling model, a 975B-parameter open-weights multimodal MoE model that reasons over text, images, and audio. It offers the same capabilities as standard Inkling at lower cost for asynchronous, non-latency-sensitive workloads.
Jul 17, 2026524k context0.8x cost rate
Kimi K3
Kimi K3 is Moonshot AI's July 2026 flagship model, a 2.8-trillion-parameter mixture-of-experts model with native multimodal vision and a 1M-token context window. It is built for long-horizon coding, reasoning, and agent workflows, positioned as an open-weight competitor to top proprietary frontier models.
Jul 16, 20261M context3x cost rate
Muse Spark 1.1
Muse Spark 1.1 is a large language model from Meta Superintelligence Labs, the first model in Meta's Muse family, designed for multimodal reasoning, coding, and AI-assisted software development with up to a million tokens of context. It targets complex agentic and coding workflows requiring long-context understanding.
Jul 16, 20261M context0.9x cost rate
KAT-Coder-Air V2.5
KAT-Coder-Air V2.5 is a lighter, faster member of Kwaipilot's KAT-Coder agentic coding model family, trained on large numbers of verifiable repository environments. It is designed for autonomous software engineering tasks like issue resolution and repo-level code changes, optimized for lower cost and latency than the Pro variant.
Jul 10, 2026256k context0.1x cost rate
KAT-Coder-Pro V2.5
KAT-Coder-Pro V2.5 is Kwaipilot's flagship agentic coding model, trained on 100,000+ verifiable repository environments and capable of autonomously handling an entire issue or business workflow end to end in a real codebase. It ranks highly on agentic coding benchmarks like PinchBench and SWE-Bench Pro.
Jul 10, 2026256k context0.6x cost rate
GPT-5.6 Luna Pro
GPT-5.6 Luna Pro is a higher-reasoning-effort version of the Luna tier in OpenAI's GPT-5.6 family, aimed at latency-sensitive chat, classification, and lightweight agentic tasks that need a bit more accuracy than the base Luna model.
Jul 9, 20261.1M context0.2x cost rate
GPT-5.6 Luna Pro (batch)
GPT-5.6 Luna Pro is a higher-reasoning-effort version of the fast, low-cost Luna tier in OpenAI's GPT-5.6 family. This is the batch variant for cheaper asynchronous processing.
Jul 9, 20261.1M context0.1x cost rate
GPT-5.6 Luna
GPT-5.6 Luna is the fast, low-cost tier of OpenAI's GPT-5.6 family, tuned for high-volume, latency-sensitive work like chat, classification, and lightweight agents while retaining genuine GPT-5.6 reasoning.
Jul 9, 20261.1M context0.2x cost rate
GPT-5.6 Luna (batch)
GPT-5.6 Luna is the fast, low-cost tier of OpenAI's GPT-5.6 family for high-volume chat, classification, and lightweight agents. This is the batch variant for cheaper asynchronous processing.
Jul 9, 20261.1M context0.1x cost rate
GPT-5.6 Terra Pro
GPT-5.6 Terra Pro is a higher-reasoning-effort version of the balanced Terra tier in OpenAI's GPT-5.6 family, for everyday coding, reasoning, and agentic tasks that need a bit more accuracy.
Jul 9, 20261.1M context2x cost rate
GPT-5.6 Terra Pro (batch)
GPT-5.6 Terra Pro is a higher-reasoning-effort version of the balanced Terra tier in OpenAI's GPT-5.6 family. This is the batch variant for cheaper asynchronous processing.
Jul 9, 20261.1M context1x cost rate
GPT-5.6 Terra
GPT-5.6 Terra is the balanced, everyday tier of OpenAI's GPT-5.6 family, offering coding, reasoning, and agentic performance competitive with GPT-5.5 at roughly half the cost.
Jul 9, 20261.1M context2x cost rate
GPT-5.6 Terra (batch)
GPT-5.6 Terra is the balanced, everyday tier of OpenAI's GPT-5.6 family, offering coding, reasoning, and agentic performance at low cost. This is the batch variant for cheaper asynchronous processing.
Jul 9, 20261.1M context1x cost rate
GPT-5.6 Sol Pro
GPT-5.6 Sol Pro is an extended-reasoning version of the Sol flagship tier in OpenAI's GPT-5.6 family, aimed at the hardest complex reasoning, multi-step coding, and long-horizon agentic tasks.
Jul 9, 20261.1M context3x cost rate
GPT-5.6 Sol Pro (batch)
GPT-5.6 Sol Pro is an extended-reasoning version of the Sol flagship tier in OpenAI's GPT-5.6 family, for the hardest reasoning and agentic coding tasks. This is the batch variant for cheaper asynchronous processing.
Jul 9, 20261.1M context1.5x cost rate
GPT-5.6 Sol
GPT-5.6 Sol is the flagship tier of OpenAI's GPT-5.6 family, built for complex reasoning, multi-step coding, and long-horizon agentic work, with particular strength on command-line and terminal-driven tasks.
Jul 9, 20261.1M context3x cost rate
GPT-5.6 Sol (batch)
GPT-5.6 Sol is the flagship tier of OpenAI's GPT-5.6 family, built for complex reasoning, multi-step coding, and long-horizon agentic work. This is the batch variant for cheaper asynchronous processing.
Jul 9, 20261.1M context1.5x cost rate
Grok 4.5
Grok 4.5 is a flagship xAI model positioned for code generation and general high-intelligence use cases, sitting between Grok 4.3 and Grok 4.6 in xAI's rapid release cadence. It is designed for demanding reasoning, coding, and agentic workloads.
Jul 8, 2026500k context1x cost rate
Grok Latest
This is an alias that always points to xAI's latest flagship Grok model.
Jul 8, 2026500k context1x cost rate
Aion-3.0-Mini
Aion-3.0-mini is a smaller, faster member of AionLabs' Aion-3.0 roleplay and storytelling family, built on DeepSeek models with a similar collaborative multi-model generation approach. It targets the same creative-writing and roleplay use cases at lower cost and latency.
Jul 7, 2026131k context0.3x cost rate
Aion-3.0
Aion-3.0 is a multi-model roleplaying and storytelling system from AionLabs built on the GLM family, using a collaborative generation process where multiple specialized models contribute to each response. It is designed for rich narrative structure and compelling character-driven fiction.
Jul 7, 2026131k context1.5x cost rate
Hy3
Hy3 is Tencent Hunyuan's open-source flagship model, a 295B-parameter MoE (about 21B active) with a hybrid fast/slow reasoning system and 256k context. It targets coding, research, and agentic product integrations, emphasizing reliable tool calling and long multi-turn conversations.
Jul 6, 2026262k context0.1x cost rate
Laguna XS 2.1
Laguna XS 2.1 is Poolside's smaller open-weight coding model (33B total parameters MoE), built for local, lightweight agentic coding workloads. It applies learnings from Poolside's larger Laguna models while staying fast and cheap enough to run on modest hardware.
Jul 2, 2026262k context0.1x cost rate
Laguna XS 2.1 (free)
Free-tier variant of Poolside's Laguna XS 2.1, a compact open-weight MoE coding model designed for local agentic coding. Same underlying model as the paid version, aimed at lightweight, cost-effective code generation and tool-use tasks.
Jul 2, 2026262k contextFree
Claude Sonnet 5
Claude Sonnet 5 is Anthropic's most capable Sonnet-tier model, bringing near-Opus intelligence to coding, agentic workflows, and professional knowledge work at Sonnet-level cost. It is a strict improvement over Sonnet 4.6 across the cost-performance curve and serves as the default model for Anthropic's Free and Pro plans.
Jun 30, 20261M context2x cost rate
Claude Sonnet 5 (batch)
Batch-processing variant of Claude Sonnet 5, Anthropic's near-Opus-capable Sonnet-tier flagship. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Jun 30, 20261M context1x cost rate
Nex-N2-Mini
Nex-N2-mini is the smaller variant of Nex AGI's Nex-N2 agentic model family, built on the Qwen3.5 architecture. It is designed for coding, tool use, and long-horizon agentic workflows with lower latency than the Pro variant.
Jun 24, 2026262k context0.1x cost rate
Fugu Ultra
Sakana Fugu Ultra is Sakana AI's multi-agent orchestration model that coordinates a pool of frontier language models through a single API endpoint, dynamically assigning roles via learned orchestration. It targets complex, multi-step tasks like agentic software engineering, achieving top-tier benchmark scores by intelligently routing work across underlying models.
Jun 24, 20261M context6x cost rate
North Mini Code (free)
North Mini Code is Cohere's first developer-focused coding model, a 30B-total/3B-active mixture-of-experts model released open-weight under Apache 2.0. It is trained for agentic coding, terminal tasks, and tool use across agent harnesses like OpenCode and SWE-Agent, with a 256K context window; this is the free-tier variant of the same model.
Jun 17, 2026256k contextFree
GLM 5.2
GLM-5.2 is Z.ai's flagship open-source model, shipping with a 1M-token context window, two selectable reasoning effort levels, and an unrestricted MIT license. It is competitive with top closed models on demanding software-engineering benchmarks, trailing Claude Opus-class models only narrowly while outperforming several rivals on coding tasks.
Jun 16, 20261M context0.7x cost rate
GLM 5.2 (batch)
This is the batch-processing variant of Z.ai's GLM-5.2 flagship model, offering the same 1M-context, MIT-licensed capabilities for coding and agentic reasoning at lower cost for asynchronous workloads.
Jun 16, 20261M context1x cost rate
GLM 5.2 (free)
This is the free OpenRouter endpoint for Z.ai's GLM-5.2 flagship model, offering the same 1M-context, MIT-licensed coding and agentic-reasoning capabilities at no token cost. Free access is rate-limited, making it best suited for experimentation and low-volume use rather than production workloads.
Jun 16, 2026256k contextFree
Kimi K2.7 Code
Kimi K2.7 Code is Moonshot AI's June 2026 specialized coding model, tuned for stronger instruction-following over long contexts and higher success on coding tasks. It cuts thinking-token usage by about 30% versus K2.6 while improving benchmark scores, making it well suited for agentic coding workflows.
Jun 12, 2026262k context0.7x cost rate
Kimi K2.7 Code (batch)
Kimi K2.7 Code is Moonshot AI's June 2026 specialized coding model, tuned for stronger instruction-following over long contexts and higher success on coding tasks, cutting thinking-token usage versus K2.6 while improving benchmarks. This is the batch variant for asynchronous, lower-cost processing.
Jun 12, 2026262k context0.8x cost rate
Claude Fable Latest
This is an alias that always points to Anthropic's latest Claude Fable model.
Jun 9, 20261M context10x cost rate
Claude Fable 5
Claude Fable 5 is Anthropic's most capable coding-focused model, built for long-running, ambiguous, asynchronous work such as legacy migrations and gnarly production bugs that can span hours or days. It self-corrects through verification loops and is state of the art on nearly all tested coding benchmarks, with a 1M-token context window.
Jun 9, 20261M context10x cost rate
Claude Fable 5 (batch)
Batch-processing variant of Claude Fable 5, Anthropic's most capable coding model for long-running, ambiguous, asynchronous engineering work. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Jun 9, 20261M context5x cost rate
Nex-N2-Pro
Nex-N2-Pro is Nex AGI's flagship agentic mixture-of-experts model with 17B active of 397B total parameters, built on the Qwen3.5 architecture and released under Apache 2.0. It unifies planning, coding, debugging, and iteration into a single agentic loop, targeting deep research and long-horizon coding tasks.
Jun 8, 2026262k context0.2x cost rate
Nemotron 3.5 Content Safety (free)
Nemotron 3.5 Content Safety is a small NVIDIA moderation model built on Google's Gemma-3-4B and fine-tuned on multimodal, multilingual safety data. It classifies prompts and responses across 23 safety categories such as violence, hate, self-harm, and PII in 12 languages, and is meant for guardrailing other models rather than general chat.
Jun 4, 2026128k contextFree
Nemotron 3 Ultra
Nemotron 3 Ultra is NVIDIA's flagship open-weight model in the Nemotron 3 family, a 550B-parameter mixture-of-experts model with 55B active parameters. It is optimized for orchestrating complex, long-running agent workflows, combining frontier reasoning with high inference throughput.
Jun 4, 2026512k context0.7x cost rate
Nemotron 3 Ultra (batch)
Nemotron 3 Ultra is NVIDIA's flagship open-weight model in the Nemotron 3 family, a 550B-parameter mixture-of-experts model with 55B active parameters optimized for complex, long-running agent workflows. This is the batch variant for asynchronous, lower-cost processing.
Jun 4, 2026512k context0.7x cost rate
Nemotron 3 Ultra (free)
Nemotron 3 Ultra is NVIDIA's flagship open-weight model in the Nemotron 3 family, a 550B-parameter mixture-of-experts model with 55B active parameters optimized for complex, long-running agent workflows. This is the free-tier variant of the same model.
Jun 4, 20261M contextFree
Qwen3.7 Plus
Qwen3.7 Plus is Alibaba's multimodal agent-focused model in the API-only Qwen3.7 generation, combining vision and language understanding with strong agentic tool-use capability.
Jun 3, 20261M context0.3x cost rate
MiniMax M3
MiniMax-M3 is MiniMax's latest M-series model for agentic reasoning, tool use, coding, multimodal chat input, and long-context tasks. It represents the newest generation of MiniMax's flagship line, building on the coding and agentic strengths of the M2 series.
May 31, 20261M context0.3x cost rate
MiniMax M3 (batch)
MiniMax M3 is an open-weight flagship model from MiniMax combining frontier-level coding performance, native multimodal image and video understanding, and a 1M-token context window. It is well suited for complex agentic coding and long-context tasks; this is the batch variant for asynchronous, lower-cost processing.
May 31, 2026524k context0.3x cost rate
Step 3.7 Flash
Step 3.7 Flash is StepFun's multimodal vision-language model, built on Step 3.5 Flash with an added vision encoder for native image understanding. With a 256k context window, selectable reasoning levels, and high throughput, it targets agentic, coding, and search workflows that mix text and visual input.
May 28, 2026262k context0.2x cost rate
Claude Opus 4.8 (Fast)
Fast-mode variant of Claude Opus 4.8, delivering Anthropic's top Opus-tier intelligence with reduced latency for demanding coding and agentic workloads. Same underlying capabilities as the standard model, tuned for quicker responses.
May 27, 20261M context10x cost rate
Claude Opus 4.8
Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family at this point in the lineup, supporting text, image, and file input with reasoning and a 1M-token context window. It is aimed at the highest-difficulty coding, analysis, and agentic tasks.
May 27, 20261M context5x cost rate
Claude Opus 4.8 (batch)
Batch-processing variant of Claude Opus 4.8, Anthropic's most capable generally available Opus model. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
May 27, 20261M context2.5x cost rate
Qwen3.7 Max
Qwen3.7 Max is Alibaba's flagship API-only model (no open weights) in the Qwen3.7 generation, positioned as the text flagship for advanced reasoning and agentic tasks.
May 21, 20261M context1x cost rate
Grok Build 0.1
Grok Build 0.1 is xAI's fast coding-specialist model, trained specifically for agentic software-engineering workflows like code generation, review, debugging, and multi-step development rather than general chat. It has a 256K context window with text and image input and is optimized for interactive coding agents and high-volume tool-use workloads.
May 20, 2026256k context0.5x cost rate
Gemini 3.5 Flash
Gemini 3.5 Flash is Google's near-Pro intelligence model at Flash-tier cost and speed, delivering strong coding proficiency and parallel agentic execution for complex, long-horizon tasks. It is designed to give developers frontier-level agent and coding performance without Pro-level pricing.
May 19, 20261M context1.5x cost rate
Gemini 3.5 Flash (batch)
Gemini 3.5 Flash is Google's near-Pro intelligence model at Flash-tier cost and speed, delivering strong coding proficiency and parallel agentic execution for complex, long-horizon tasks. This is the batch variant for asynchronous, lower-cost processing.
May 19, 20261M context0.9x cost rate
Claude Opus 4.7 (Fast)
Fast-mode variant of Claude Opus 4.7, trading a small amount of latency-optimized throughput for the same long-running agentic coding and reasoning capability as the standard model. Suited to workflows that need Opus-level intelligence with quicker turnaround.
May 12, 20261M context30x cost rate
Perceptron Mk1
Perceptron Mk1 is the flagship vision-language model from Perceptron AI, a startup founded by former Meta FAIR researchers, built for video understanding and embodied/spatial reasoning. It is designed for video QA, summarization and event detection, OCR and document parsing on messy real-world inputs, object detection/counting, and pose estimation, delivering frontier-competitive results at a much lower cost.
May 12, 202633k context0.3x cost rate
Ring-2.6-1T
Ring 2.6 1T is a trillion-parameter open-weight reasoning model from InclusionAI (Ant Group), the deep-thinking counterpart built on top of the Ling base for complex workflows. It is purpose-built for agentic coding, tool use, and long-horizon multi-step task execution, with benchmark results surpassing GPT-5.4 and Gemini 3.1 Pro on some agentic tasks.
May 8, 2026262k context0.1x cost rate
Gemini 3.1 Flash Lite
Gemini 3.1 Flash-Lite is Google's high-efficiency multimodal model optimized for low-latency, high-volume workloads across text, image, video, audio, and PDF input. It is designed for lightweight agentic workflows and simple data extraction where responsiveness and cost matter most.
May 7, 20261M context0.3x cost rate
Gemini 3.1 Flash Lite (batch)
Gemini 3.1 Flash-Lite is Google's high-efficiency multimodal model optimized for low-latency, high-volume workloads across text, image, video, audio, and PDF input. This is the batch variant for asynchronous, lower-cost processing.
May 7, 20261M context0.1x cost rate
GPT Chat Latest
GPT Chat Latest is an alias that always points to the current production ChatGPT model snapshot, tuned for natural conversational chat rather than heavy extended reasoning. It suits general assistant use where staying in sync with ChatGPT's live behavior matters.
May 5, 2026400k context6x cost rate
Grok 4.3
Grok 4.3 is an iteration in xAI's Grok 4.x line, improving on Grok 4.20 with incremental gains in reasoning, coding, and agentic task performance. It is positioned as a high-intelligence general-purpose model for complex problem solving.
Apr 30, 20261M context0.6x cost rate
Granite 4.1 8B
Granite 4.1 8B is IBM's dense, long-context instruct model with an enhanced post-training pipeline for improved tool calling, instruction following, and coding. It matches or outperforms IBM's larger Granite 4.0 32B MoE model despite its smaller size, and is released under Apache 2.0.
Apr 30, 2026131k context0.1x cost rate
Mistral Medium 3.5
Mistral Medium 3.5 is Mistral AI's April 2026 flagship-tier model, a dense 128B model with a 256K context window that merges instruction-following, reasoning, and coding into a single set of weights. It is the recommended successor to Mistral Medium 3.1 for general-purpose and agentic tasks.
Apr 30, 2026262k context1.5x cost rate
Nemotron 3 Nano Omni (free)
Nemotron 3 Nano Omni Reasoning is a multimodal, reasoning-tuned variant of NVIDIA's Nemotron 3 Nano, built for omni (text and image, likely audio) understanding combined with extended chain-of-thought. It targets efficient agentic applications that need both reasoning and multimodal input at low compute cost.
Apr 28, 2026256k contextFree
Anthropic Claude Haiku Latest
This is an alias that always points to Anthropic's latest Claude Haiku model.
Apr 27, 2026200k context1x cost rate
OpenAI GPT Mini Latest
This is an alias that always points to OpenAI's latest smaller/cheaper GPT model.
Apr 27, 2026400k context0.9x cost rate
Google Gemini Pro Latest
This is an alias that always points to Google's latest Gemini Pro model.
Apr 27, 20261M context2x cost rate
MoonshotAI Kimi Latest
This is an alias that always points to Moonshot AI's latest Kimi model.
Apr 27, 20261M context2.5x cost rate
Google Gemini Flash Latest
This is an alias that always points to Google's latest Gemini Flash model.
Apr 27, 20261M context0.4x cost rate
Anthropic Claude Sonnet Latest
This is an alias that always points to Anthropic's latest Claude Sonnet model.
Apr 27, 20261M context2x cost rate
OpenAI GPT Latest
This is an alias that always points to OpenAI's latest flagship GPT model.
Apr 27, 20261.1M context3x cost rate
Qwen3.5 Plus 2026-04-20
An April 2026 snapshot of Qwen3.5 Plus, Alibaba's proprietary mid-tier model in the Qwen3.5 generation, offering the same general-purpose reasoning and agentic capability as other Plus-tier Qwen3.5 releases.
Apr 27, 20261M context0.3x cost rate
Qwen3.6 Flash
Qwen3.6 Flash is Alibaba's fast, low-cost proprietary tier of the Qwen3.6 generation, aimed at high-throughput, latency-sensitive chat and agentic tasks.
Apr 27, 20261M context0.2x cost rate
Qwen3.6 35B A3B
Qwen3.6 35B-A3B is a sparse mixture-of-experts model from Alibaba's Qwen3.6 line (35B total, 3B active parameters), offering efficient general-purpose and agentic capability at low inference cost.
Apr 27, 2026262k context0.2x cost rate
Qwen3.6 Max Preview
Qwen3.6 Max Preview is a preview of Alibaba's flagship proprietary Qwen3.6 model, intended for top-tier reasoning, coding, and agentic performance ahead of general availability.
Apr 27, 2026262k context1x cost rate
Qwen3.6 27B
Qwen3.6 27B is an open-weight model from Alibaba's Qwen3.6 generation, a successor to Qwen3.5 with incremental improvements to general chat, reasoning, and agentic performance.
Apr 27, 2026262k context0.7x cost rate
GPT-5.5 Pro
GPT-5.5 Pro is the extended-reasoning tier of GPT-5.5, for the hardest coding, research, and analytical tasks where maximum accuracy is worth the extra latency and cost.
Apr 24, 20261.1M context35x cost rate
GPT-5.5 Pro (batch)
GPT-5.5 Pro is the extended-reasoning tier of GPT-5.5 for the hardest coding, research, and analytical tasks. This is the batch variant for cheaper asynchronous processing.
Apr 24, 20261.1M context18x cost rate
GPT-5.5
GPT-5.5 is OpenAI's April 2026 flagship model, described as its most intuitive yet at understanding intent and carrying more of a task itself. It excels at writing and debugging code, research, data analysis, and multi-step agentic tool use, with reduced hallucination in sensitive domains like law, medicine, and finance.
Apr 24, 20261.1M context6x cost rate
GPT-5.5 (batch)
GPT-5.5 is OpenAI's April 2026 flagship model, excelling at coding, research, data analysis, and agentic tool use with reduced hallucination in sensitive domains. This is the batch variant for cheaper asynchronous processing.
Apr 24, 20261.1M context3x cost rate
DeepSeek V4 Pro 0423
DeepSeek V4 Pro is the flagship variant of DeepSeek's V4 model family, featuring reasoning-effort levels (low/high/max) and strong performance across coding, reasoning, and agentic benchmarks. It is DeepSeek's top-tier general-purpose and reasoning model in the V4 generation.
Apr 24, 20261M context0.8x cost rate
DeepSeek V4 Flash 0423
DeepSeek V4 Flash is the faster, cheaper variant of DeepSeek's V4 model family, released in public beta and reported to substantially exceed V4-Pro-Preview on agent benchmarks. It is designed for high-throughput agentic and coding workloads where speed and cost matter alongside strong capability.
Apr 24, 20261M context0.1x cost rate
Ling-2.6-1T
Ling 2.6 1T is a trillion-parameter-scale open-weight Mixture-of-Experts model from InclusionAI (Ant Group), part of the fast general-purpose Ling line built for high-throughput production use. It targets large-scale reasoning, coding, and agentic workflows.
Apr 23, 2026262k context0.1x cost rate
Hy3 preview
Hy3-preview is the early-access preview release of Tencent Hunyuan's Hy3 model, a large MoE architecture with hybrid fast/slow reasoning aimed at coding, research, and agentic workflows. It offers the same general capabilities as the final Hy3 release under a more restrictive preview license.
Apr 22, 2026262k context0.1x cost rate
MiMo-V2.5-Pro
MiMo v2.5 Pro is the larger, higher-capability variant of Xiaomi's MiMo v2.5 open-source reasoning model line. It targets more demanding reasoning, math, and coding tasks than the base MiMo v2.5 model while retaining an open-source, efficiency-conscious design.
Apr 22, 20261.1M context0.2x cost rate
MiMo-V2.5
MiMo v2.5 is part of Xiaomi's MiMo family of open-source reasoning models, following the original MiMo-7B focused on math, code, and reasoning-heavy tasks. It scales up Xiaomi's reasoning-focused training approach for stronger multimodal and general reasoning performance.
Apr 22, 20261.1M context0.1x cost rate
Ling-2.6-flash
Ling 2.6 Flash is a high-efficiency open-weight instruct model from InclusionAI (Ant Group) with 104B total but only ~7.4B active parameters via Mixture-of-Experts, and a 262K-token context window. It is purpose-built for agentic workflows, coding, and document processing at low cost.
Apr 21, 2026262k context0.1x cost rate
Claude Opus Latest
This is an alias that always points to Anthropic's latest Claude Opus model.
Apr 21, 20261M context5x cost rate
Kimi K2.6
Kimi K2.6 is Moonshot AI's April 2026 open-weight model that ties GPT-5.5 on coding benchmarks like SWE-Bench Pro while costing roughly 80% less per token. It is designed for high-end coding, reasoning, and agentic tasks at a fraction of proprietary-model cost.
Apr 20, 2026262k context0.5x cost rate
Claude Opus 4.7
Claude Opus 4.7 is the next generation of Anthropic's Opus family, purpose-built for long-running, asynchronous agents that operate with minimal supervision. It carries a 1M-token context window and targets complex, multi-step engineering and research tasks.
Apr 16, 20261M context5x cost rate
Claude Opus 4.7 (batch)
Batch-processing variant of Claude Opus 4.7, Anthropic's Opus model built for long-running asynchronous agents. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Apr 16, 20261M context2.5x cost rate
GLM 5.1
GLM-5.1 is an incremental upgrade to Z.ai's GLM-5 flagship model, released as open source in April 2026 with improved reasoning and agentic performance. It continues GLM-5's focus on complex systems design and large-scale programming tasks.
Apr 7, 2026205k context0.7x cost rate
Gemma 4 26B A4B
Gemma 4 26B-A4B IT is a Mixture-of-Experts model from Google's open-weight Gemma 4 family with 26B total but only ~4B active parameters, giving high-throughput inference for advanced reasoning tasks. It is a multimodal, hybrid-thinking model supporting 140+ languages and up to 256K context.
Apr 3, 2026262k context0.1x cost rate
Gemma 4 26B A4B (free)
Gemma 4 26B-A4B IT is a Mixture-of-Experts model from Google's open-weight Gemma 4 family with 26B total but only ~4B active parameters, giving high-throughput inference for advanced reasoning tasks. This is the free-tier variant of the same model.
Apr 3, 2026262k contextFree
Gemma 4 31B
Gemma 4 31B IT is a dense model from Google's open-weight Gemma 4 family that bridges server-grade performance and local execution, ranking near the top of open-model leaderboards. It handles multimodal, hybrid-thinking reasoning across 140+ languages with up to 256K context, well suited for advanced chat, coding, and reasoning.
Apr 2, 2026262k context0.1x cost rate
Gemma 4 31B (free)
Gemma 4 31B IT is a dense model from Google's open-weight Gemma 4 family that bridges server-grade performance and local execution, ranking near the top of open-model leaderboards. This is the free-tier variant of the same model.
Apr 2, 2026262k contextFree
Qwen3.6 Plus
Qwen3.6 Plus is Alibaba's proprietary mid-tier model in the Qwen3.6 generation, balancing cost and capability for general chat, reasoning, and agentic use cases.
Apr 2, 20261M context0.4x cost rate
GLM 5V Turbo
GLM-5V-Turbo is a fast, lower-latency vision-language variant in Z.ai's GLM-5 family, combining multimodal image/video understanding with efficient serving. It is designed for agentic and multimodal workflows where speed and cost matter alongside visual reasoning.
Apr 1, 2026203k context0.9x cost rate
Trinity Large Thinking
Trinity-Large-Thinking is Arcee AI's open-source frontier reasoning model, a 399B-parameter sparse mixture-of-experts model (about 13B active per token) with a 512K-token context window, released under Apache 2.0. It generates explicit chain-of-thought reasoning traces and is built for complex, long-horizon agentic tasks and long-document analysis.
Apr 1, 2026262k context0.2x cost rate
Grok 4.20 Multi-Agent
Grok 4.20 Multi-Agent is a variant of xAI's Grok 4.20 that runs several specialized agents on a shared backbone within a single API call to perform deep, multi-step research. It combines web search, code execution, and data synthesis tools to deliver comprehensive, well-sourced answers for research-heavy tasks.
Mar 31, 20262M context0.6x cost rate
Grok 4.20
Grok 4.20 is an xAI flagship model built for long-running agentic work and ambitious interactive and visual tasks, with a 500,000-token context window and text/image input. It targets high-intelligence coding, reasoning, and multi-step agent use cases.
Mar 31, 20262M context0.6x cost rate
Lyria 3 Pro Preview
Lyria 3 Pro is Google DeepMind's flagship music generation model, capable of producing full compositions up to three minutes long with structural awareness of intros, verses, choruses, and bridges. It supports multi-language vocals and genre versatility for professional-grade, prompt-driven music creation.
Mar 30, 20261M contextFree
Lyria 3 Clip Preview
Lyria 3 (clip preview) is Google DeepMind's music generation model that produces high-fidelity stereo audio clips up to 30 seconds from text prompts. It is designed for rapid prototyping, social media assets, and short-form audio generation rather than full song composition.
Mar 30, 20261M contextFree
KAT-Coder-Pro V2
KAT-Coder-Pro V2 is an agentic coding model from Kwaipilot (Kuaishou) designed to autonomously locate and modify code across entire real, executable repositories rather than produce single-turn snippets. It targets professional software engineering workflows such as issue resolution and multi-step repo changes.
Mar 27, 2026262k context0.3x cost rate
Reka Edge
Reka Edge is a small, efficient model from Reka AI designed for low-latency, on-device or resource-constrained deployments. It targets simple chat and task completion where speed and low cost outweigh the need for top-end capability.
Mar 20, 202616k context0.1x cost rate
MiniMax M2.7
MiniMax-M2.7 is MiniMax's model for software engineering, agent workflows, and professional document tasks, notable as a self-evolving model that ran autonomous reinforcement-learning optimization loops during training. It targets advanced coding and long-horizon agentic use cases.
Mar 18, 2026205k context0.3x cost rate
GPT-5.4 Nano
GPT-5.4 Nano is the smallest, fastest model in the GPT-5.4 family, aimed at high-throughput, low-latency tasks like classification and simple extraction where cost efficiency is the priority.
Mar 17, 2026400k context0.2x cost rate
GPT-5.4 Nano (batch)
GPT-5.4 Nano is the smallest, fastest model in the GPT-5.4 family, aimed at high-throughput, low-latency tasks. This is the batch variant for cheaper asynchronous processing.
Mar 17, 2026400k context0.1x cost rate
GPT-5.4 Mini
GPT-5.4 Mini is the smaller, cheaper counterpart to GPT-5.4, suited to everyday chat, lightweight coding, and agentic tasks where speed and cost outweigh maximum reasoning depth.
Mar 17, 2026400k context0.9x cost rate
GPT-5.4 Mini (batch)
GPT-5.4 Mini is the smaller, cheaper counterpart to GPT-5.4 for everyday chat, lightweight coding, and agentic tasks. This is the batch variant for cheaper asynchronous processing.
Mar 17, 2026400k context0.4x cost rate
Mistral Small 4
Mistral Small 4 (2603) is Mistral AI's March 2026 model, a 119B-parameter mixture-of-experts model with 6B active parameters that unifies instruct, reasoning, and coding with a toggleable reasoning mode and 256K context. It combines reasoning strength from Magistral, vision from Pixtral, and coding ability from Devstral in one efficient model.
Mar 16, 2026262k context0.1x cost rate
GLM 5 Turbo
GLM-5-Turbo is a faster, lower-latency serving variant of Z.ai's flagship GLM-5 model, aimed at reducing cost and response time while retaining GLM-5's strengths in coding and long-horizon agentic workflows.
Mar 15, 2026203k context0.9x cost rate
Nemotron 3 Super
Nemotron 3 Super is a 120B mixture-of-experts model with 12B active parameters in NVIDIA's Nemotron 3 family, sitting between Nano and Ultra in scale. It offers stronger reasoning and agentic capability than Nemotron 3 Nano while remaining more efficient than Nemotron 3 Ultra.
Mar 11, 20261M context0.1x cost rate
Nemotron 3 Super (free)
Nemotron 3 Super is a 120B mixture-of-experts model with 12B active parameters in NVIDIA's Nemotron 3 family, offering stronger reasoning and agentic capability than Nemotron 3 Nano while staying more efficient than Nemotron 3 Ultra. This is the free-tier variant of the same model.
Mar 11, 2026262k contextFree
Seed-2.0-Lite
Seed 2.0 Lite is a cost-efficient, low-latency multimodal model from ByteDance Seed, delivering solid text, vision, and agentic tool-use capability. It is designed as a practical default for high-frequency production workloads.
Mar 10, 2026262k context0.4x cost rate
Qwen3.5-9B
Qwen3.5 9B is a small open-weight model from Alibaba's Qwen3.5 generation, suited for fast, low-cost general chat and simple coding tasks with the efficiency gains of the 3.5 series.
Mar 10, 2026262k context0.1x cost rate
GPT-5.4 Pro
GPT-5.4 Pro is the extended-reasoning tier of GPT-5.4, prized for low hallucination rates on hard problems. It suits complex research, coding, and analysis where reliability matters more than speed.
Mar 5, 20261.1M context35x cost rate
GPT-5.4 Pro (batch)
GPT-5.4 Pro is the extended-reasoning tier of GPT-5.4 for complex research, coding, and analysis that need reliable, low-hallucination answers. This is the batch variant for cheaper asynchronous processing.
Mar 5, 20261.1M context18x cost rate
GPT-5.4
GPT-5.4 is OpenAI's March 2026 flagship model, released in Thinking and Pro variants known for low hallucination rates. It targets advanced reasoning, research, and coding tasks that need reliable, well-grounded answers.
Mar 5, 20261.1M context3x cost rate
GPT-5.4 (batch)
GPT-5.4 is OpenAI's March 2026 flagship model known for low hallucination rates, suited to advanced reasoning, research, and coding. This is the batch variant for cheaper asynchronous processing.
Mar 5, 20261.1M context1.5x cost rate
Mercury 2
Mercury 2 is Inception Labs' diffusion-based LLM that generates responses in parallel rather than token-by-token, delivering very high throughput (1000+ tokens/second) with a 128K context window. It offers tunable reasoning and native tool use, matching Claude Haiku/Gemini Flash-class quality at much faster speeds and lower cost, ideal for latency-sensitive agentic and coding tasks.
Mar 4, 2026128k context0.2x cost rate
Gemini 3.1 Flash Lite Preview
Gemini 3.1 Flash-Lite Preview is the preview release of Google's high-efficiency multimodal model, optimized for low-latency, high-volume workloads such as lightweight agentic workflows and simple data extraction. It offers the same responsiveness and cost focus as the GA Flash-Lite release.
Mar 3, 20261M context0.3x cost rate
Seed-2.0-Mini
Seed 2.0 Mini is ByteDance Seed's smallest 2.0-generation model, targeting latency-sensitive, high-concurrency, cost-sensitive scenarios with a 256K context window and four reasoning-effort modes. It delivers performance comparable to Seed 1.6 at lower cost.
Feb 26, 2026262k context0.1x cost rate
Qwen3.5-35B-A3B
Qwen3.5 35B-A3B is a sparse mixture-of-experts model from Alibaba's Qwen3.5 line (35B total, 3B active parameters), built for efficient general-purpose and agentic tasks at low inference cost.
Feb 25, 2026262k context0.3x cost rate
Qwen3.5-27B
Qwen3.5 27B is a mid-size dense open-weight model from Alibaba's Qwen3.5 generation, offering general-purpose chat, coding, and agentic capability with efficiency improvements over the Qwen3 series.
Feb 25, 2026262k context0.3x cost rate
Qwen3.5-122B-A10B
Qwen3.5 122B-A10B is one of Alibaba's Qwen3.5 open-weight mixture-of-experts models (122B total, 10B active), part of a generation positioned for the 'agentic AI era' with improved tool use and multi-step task handling over Qwen3.
Feb 25, 2026262k context0.4x cost rate
Qwen3.5-Flash
Qwen3.5 Flash is Alibaba's fast, low-cost proprietary tier of the Qwen3.5 generation (this snapshot dated February 2026), designed for high-throughput, latency-sensitive general and agentic tasks.
Feb 25, 20261M context0.1x cost rate
Gemini 3.1 Pro Preview Custom Tools
Gemini 3.1 Pro Preview (custom tools) is a variant of Google's Gemini 3.1 Pro Preview configured for custom tool-calling setups, sharing the base model's strengths in agentic coding, long-horizon tool orchestration, structured planning, and multimodal analysis.
Feb 25, 20261M context2x cost rate
GPT-5.3-Codex
GPT-5.3-Codex is OpenAI's February 2026 Codex-native coding agent, pairing frontier coding performance with general reasoning for long-horizon, real-world engineering work. It supports steering the agent mid-task and is notable as a model OpenAI used to help build itself.
Feb 24, 2026400k context2.5x cost rate
Aion-2.0
Aion-2.0 is a variant of DeepSeek V3.2 from AionLabs, tuned specifically for immersive roleplaying and collaborative storytelling. It is particularly strong at introducing tension, conflict, and mature or darker themes with nuance and depth.
Feb 23, 2026131k context0.4x cost rate
Gemini 3.1 Pro Preview
Gemini 3.1 Pro Preview is Google's advanced reasoning model improving long-horizon stability and tool orchestration over Gemini 3 Pro, with a new medium thinking level to balance cost, speed, and performance. It excels at agentic coding, structured planning, multimodal analysis, and workflow automation for autonomous agents and enterprise tasks.
Feb 19, 20261M context2x cost rate
Gemini 3.1 Pro Preview (batch)
Gemini 3.1 Pro Preview is Google's advanced reasoning model improving long-horizon stability and tool orchestration, well suited to agentic coding, structured planning, multimodal analysis, and workflow automation. This is the batch variant for asynchronous, lower-cost processing.
Feb 19, 20261M context1x cost rate
Claude Sonnet 4.6
Claude Sonnet 4.6 is a further refined mid-tier model, engineered for complex enterprise workflows, deep software engineering, and autonomous agentic execution with a 1M-token context window in beta. It scores strongly on coding (SWE-bench) and computer-use benchmarks while keeping Sonnet-tier pricing.
Feb 17, 20261M context3x cost rate
Claude Sonnet 4.6 (batch)
Batch-processing variant of Claude Sonnet 4.6, Anthropic's mid-tier model for enterprise coding and agentic execution. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Feb 17, 20261M context1.5x cost rate
Qwen3.5 Plus 2026-02-15
A February 2026 snapshot of Qwen3.5 Plus, Alibaba's proprietary mid-tier Qwen3.5 model, balancing cost and capability for general chat, reasoning, and agentic workloads.
Feb 16, 20261M context0.3x cost rate
Qwen3.5 397B A17B
Qwen3.5 397B-A17B is the largest and first-released model in Alibaba's Qwen3.5 mixture-of-experts generation, targeting flagship-level reasoning, coding, and agentic performance among open-weight models.
Feb 16, 2026262k context0.5x cost rate
MiniMax M2.5
MiniMax-M2.5 is a further update in MiniMax's M2 series, achieving state-of-the-art results among MiniMax models in programming, tool calling, search, and office productivity tasks. It is aimed at real-world developer and knowledge-work productivity.
Feb 12, 2026205k context0.2x cost rate
GLM 5
GLM-5 is Z.ai's flagship open-source foundation model, scaling to 744B total parameters (40B active) and trained on 28.5T tokens. It is engineered for complex systems design and long-horizon agent workflows, delivering production-grade performance on large-scale programming tasks for expert developers.
Feb 11, 2026205k context0.4x cost rate
Qwen3 Max Thinking
The extended chain-of-thought variant of Qwen3 Max, Alibaba's flagship model, built for the hardest reasoning, math, and multi-step agentic tasks where deliberate step-by-step thinking improves accuracy.
Feb 9, 2026262k context0.8x cost rate
Claude Opus 4.6
Claude Opus 4.6 is Anthropic's strongest model for coding and long-running professional tasks in this generation, with a 1M-token context window and 128K max output. It is designed for the most demanding software engineering and multi-step agentic workflows.
Feb 4, 20261M context5x cost rate
Claude Opus 4.6 (batch)
Batch-processing variant of Claude Opus 4.6, Anthropic's strongest coding-focused Opus model for long-running professional tasks. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Feb 4, 20261M context2.5x cost rate
Qwen3 Coder Next
Qwen3 Coder Next is a next-generation iteration of Alibaba's Qwen3 Coder line, likely built on the more efficient Qwen3-Next architecture, aimed at agentic coding with improved efficiency and long-context repository handling.
Feb 4, 2026262k context0.2x cost rate
Free Models Router
OpenRouter Free is a meta-routing model from OpenRouter that automatically selects among available free-tier models to answer a request. It suits experimentation, low-stakes tasks, and cost-free testing rather than production workloads needing a specific model's guarantees.
Feb 1, 2026200k contextFree
Step 3.5 Flash
Step 3.5 Flash is StepFun's flagship sparse MoE reasoning model (196B parameters, 11B active) built for fast, reliable execution of complex tasks. It excels at decomposing and planning multi-step problems and orchestrating tool calls, making it well suited for agentic and coding workflows.
Jan 29, 2026262k context0.1x cost rate
Kimi K2.5
Kimi K2.5 is Moonshot AI's January 2026 update to Kimi K2, a 1T-parameter mixture-of-experts model with 32B active parameters and multimodal vision-language understanding, offering both instant and thinking modes. It is built for advanced agentic and conversational workflows.
Jan 27, 2026262k context0.5x cost rate
Solar Pro 3
Solar Pro 3 is Upstage's efficient large language model, building on Solar Pro 2 with Upstage's SnapPO reinforcement learning framework to sharpen step-by-step reasoning in math, code, and agentic tasks. It roughly doubles agentic benchmark performance over its predecessor while keeping the same API and serving footprint, with strong multilingual (notably Korean) support.
Jan 27, 2026131k context0.1x cost rate
MiniMax M2-her
MiniMax-M2-her is a dialogue-first model built for immersive roleplay, character-driven chat, and expressive multi-turn conversation, distilled from years of MiniMax's role-play optimization work. It powers MiniMax's Talkie companion app and is suited to storytelling and companion-style interactions rather than coding or agentic tasks.
Jan 23, 202666k context0.3x cost rate
Palmyra X5
Palmyra X5 is Writer's most advanced enterprise foundation model, built for building and scaling AI agents with a 1M-token context window and a hybrid attention architecture. It offers advanced reasoning, multi-step tool-calling, built-in RAG, and multilingual support tailored for enterprise workflows.
Jan 21, 20261M context1x cost rate
GPT Audio
GPT-Audio is OpenAI's model for natural speech input and output, generating expressive spoken responses directly from audio prompts. It is designed for voice assistants, spoken dialogue apps, and other audio-native interactions.
Jan 19, 2026128k context2x cost rate
GPT Audio Mini
GPT-Audio Mini is a smaller, cheaper version of OpenAI's GPT-Audio model for speech input and output, suited to cost-sensitive or high-volume voice assistant and spoken dialogue applications.
Jan 19, 2026128k context0.5x cost rate
GLM 4.7 Flash
GLM-4.7-Flash is a 30B-class SOTA model from Z.ai that balances performance and efficiency, offering a faster, cheaper alternative to the full GLM-4.7 model. It targets general reasoning and coding tasks where speed and cost matter more than maximum capability.
Jan 19, 2026203k context0.1x cost rate
GPT-5.2-Codex
GPT-5.2-Codex is the coding-specialized variant of GPT-5.2, built for agentic software engineering across large codebases. It benefits from GPT-5.2's large context window for repository-scale refactors and long-horizon coding tasks.
Jan 14, 2026400k context2.5x cost rate
Seed 1.6 Flash
Seed 1.6 Flash is a faster, lower-latency variant of ByteDance Seed's Seed 1.6 multimodal model, trading some depth for speed and cost efficiency. It suits high-throughput text and vision tasks where responsiveness matters most.
Dec 23, 2025262k context0.1x cost rate
Seed 1.6
Seed 1.6 is ByteDance Seed's general-purpose multimodal model with adaptive deep-thinking capability and a 256K-token context window. It is designed as a versatile foundation model for text, vision, and reasoning tasks.
Dec 23, 2025262k context0.4x cost rate
MiniMax M2.1
MiniMax-M2.1 is an update to MiniMax's M2 coding and agentic model line, focused on polyglot programming mastery and precision code refactoring. It targets developers needing strong multi-language coding assistance and reliable agentic tool use.
Dec 23, 2025205k context0.3x cost rate
GLM 4.7
GLM-4.7 is Z.ai's open-source model generation released after GLM-4.6, continuing incremental improvements in reasoning and coding. It serves as the base for the GLM-4.7-Flash variant that balances performance and efficiency at a smaller 30B class size.
Dec 22, 2025205k context0.4x cost rate
Gemini 3 Flash Preview
Gemini 3 Flash Preview is Google's fast workhorse model in the Gemini 3 line, tuned for agentic workflows, coding, and complex multi-step reasoning at low latency. It became the default model in the Gemini app for everyday responsive tasks.
Dec 17, 20251M context0.6x cost rate
Gemini 3 Flash Preview (batch)
Gemini 3 Flash Preview is Google's fast workhorse model in the Gemini 3 line, tuned for agentic workflows, coding, and complex multi-step reasoning at low latency. This is the batch variant for asynchronous, lower-cost processing.
Dec 17, 20251M context0.3x cost rate
Nemotron 3 Nano 30B A3B
Nemotron 3 Nano is NVIDIA's compute-efficient open model in the Nemotron 3 family, a 30B mixture-of-experts model with 3B active parameters. It is optimized for high-throughput, low-cost agentic and multi-agent workloads.
Dec 14, 2025262k context0.1x cost rate
Nemotron 3 Nano 30B A3B (free)
Nemotron 3 Nano is NVIDIA's compute-efficient open model in the Nemotron 3 family, a 30B mixture-of-experts model with 3B active parameters optimized for high-throughput, low-cost agentic and multi-agent workloads. This is the free-tier variant of the same model.
Dec 14, 2025256k contextFree
GPT-5.2 Chat
GPT-5.2 Chat is the fast, conversational (instant) variant of GPT-5.2, tuned for responsive everyday chat rather than deep extended reasoning. It suits general assistant use cases where low latency matters most.
Dec 10, 2025128k context2.5x cost rate
GPT-5.2 Pro
GPT-5.2 Pro is the extended-reasoning tier of GPT-5.2, trading latency for maximum accuracy on hard problems. It is aimed at complex research, coding, and analytical work with OpenAI's largest available context window.
Dec 10, 2025400k context32x cost rate
GPT-5.2 Pro (batch)
GPT-5.2 Pro is the extended-reasoning tier of GPT-5.2 for complex research, coding, and analytical work. This is the batch variant for cheaper asynchronous processing.
Dec 10, 2025400k context16x cost rate
GPT-5.2
GPT-5.2 is OpenAI's December 2025 flagship release, offered in instant, thinking, and pro modes with a 400K-token context window. It is designed for enterprise-scale reasoning, document and codebase analysis, and complex agentic workflows.
Dec 10, 2025400k context2.5x cost rate
GPT-5.2 (batch)
GPT-5.2 is OpenAI's December 2025 flagship release with a 400K-token context window, suited to enterprise-scale reasoning, document analysis, and agentic workflows. This is the batch variant for cheaper asynchronous processing.
Dec 10, 2025400k context1x cost rate
Relace Search
Relace Search is a companion model to Relace's code-apply system, built for fast retrieval of relevant code context and locations within a codebase. It is designed for AI coding agents that need quick, accurate code search before generating or applying edits.
Dec 8, 2025256k context0.7x cost rate
GLM 4.6V
GLM-4.6V is Z.ai's vision-language model that treats images, video, and tools as first-class agent inputs, extending training context to 128K tokens with native multimodal function calling. It is designed for agentic multimodal workflows such as screenshot- and document-driven tool use.
Dec 8, 2025131k context0.2x cost rate
GPT-5.1-Codex-Max
GPT-5.1-Codex-Max is OpenAI's frontier agentic coding model, using a compaction technique to coherently work across millions of tokens in a single task. It targets project-scale refactors, extended debugging sessions, and multi-hour autonomous coding loops.
Dec 4, 2025400k context2x cost rate
Nova 2 Lite
Amazon Nova 2 Lite is a fast, cost-effective multimodal reasoning model from AWS that handles text, image, video, and document input with a 1M-token context window. It supports optional extended thinking with adjustable reasoning budgets plus built-in web grounding and code interpreter tools, aimed at everyday production workloads that need quick, affordable responses.
Dec 2, 20251M context0.5x cost rate
Ministral 3 14B 2512
Ministral 3 14B is Mistral AI's December 2025 edge-focused model, offered in base, instruct, and reasoning variants with a 256K context window. It targets on-device and edge deployment with function calling and multilingual support, offering the strongest quality of the Ministral 3 sizes.
Dec 2, 2025262k context0.1x cost rate
Ministral 3 8B 2512
Ministral 3 8B is the mid-sized model in Mistral AI's December 2025 Ministral 3 edge family, balancing capability and efficiency for on-device deployment. It supports a 256K context window, function calling, and multilingual chat.
Dec 2, 2025262k context0.1x cost rate
Ministral 3 3B 2512
Ministral 3 3B is the smallest model in Mistral AI's December 2025 Ministral 3 edge family, designed for lightweight on-device deployment. It trades some capability for a tiny footprint, suiting simple chat, classification, and function-calling tasks on constrained hardware.
Dec 2, 2025131k context0.1x cost rate
Mistral Large 3 2512
Mistral Large 3 (2512) is Mistral AI's December 2025 flagship model, a sparse mixture-of-experts architecture with 41B active of 675B total parameters, released under Apache 2.0. It is Mistral's most capable model to date for advanced reasoning, coding, and multilingual enterprise workloads.
Dec 1, 2025262k context0.3x cost rate
DeepSeek V3.2
DeepSeek V3.2 is DeepSeek's V3-generation model upgrade underlying both the deepseek-chat (non-thinking) and deepseek-reasoner (thinking) modes, improving efficiency and reasoning quality over V3.1. It is designed as a strong, low-cost general-purpose and reasoning model.
Dec 1, 2025164k context0.1x cost rate
Claude Opus 4.5
Claude Opus 4.5 is a further refinement of Anthropic's Opus flagship line, with improved coding, reasoning, and long-horizon agentic capabilities over Opus 4.1. It targets the most demanding professional software engineering and analysis tasks.
Nov 24, 2025200k context5x cost rate
Claude Opus 4.5 (batch)
Batch-processing variant of Claude Opus 4.5, Anthropic's refined Opus flagship for demanding coding and reasoning work. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Nov 24, 2025200k context2.5x cost rate
Olmo 3 32B Think
OLMo 3 32B Think is Allen Institute for AI's fully open 32B reasoning model, released with open weights, training data, and code so its chain-of-thought can be inspected and traced back to training decisions. It is tuned for math and coding reasoning via long chain-of-thought, though it lacks tool use or vision support, making it best suited for transparent, auditable reasoning tasks rather than agentic workflows.
Nov 21, 202566k context0.1x cost rate
Cogito v2.1 671B
Cogito v2.1 671B is Deep Cogito's large mixture-of-experts model (671B total, 37B active parameters), trained with reinforcement learning via self-play. It features a hybrid reasoning mode achieving DeepSeek R1-level results with roughly 60% fewer reasoning tokens, suited for coding, complex reasoning, and creative writing.
Nov 13, 2025128k context0.4x cost rate
GPT-5.1
GPT-5.1 is OpenAI's late-2025 successor to GPT-5, improving conversational warmth, instruction following, and reasoning quality. It suits general-purpose chat, coding, and agentic workflows that need more natural, steerable responses.
Nov 13, 2025400k context2x cost rate
GPT-5.1 (batch)
GPT-5.1 is OpenAI's late-2025 successor to GPT-5, improving conversational warmth, instruction following, and reasoning quality for chat, coding, and agentic workflows. This is the batch variant for cheaper asynchronous processing.
Nov 13, 2025400k context0.9x cost rate
GPT-5.1-Codex
GPT-5.1-Codex is a coding-focused variant of GPT-5.1 optimized for long-running, agentic software engineering tasks in Codex or Codex-like harnesses. It is well suited to multi-step refactors, debugging, and repository-scale coding work.
Nov 13, 2025400k context2x cost rate
GPT-5.1-Codex-Mini
GPT-5.1-Codex-Mini is a smaller, cheaper version of GPT-5.1-Codex for agentic coding tasks. It suits lighter coding workflows where speed and cost matter more than the largest context or maximum autonomy.
Nov 13, 2025400k context0.4x cost rate
Kimi K2 Thinking
Kimi K2 Thinking is a reasoning-focused variant of Moonshot AI's Kimi K2 that produces extended chain-of-thought before answering. It is designed for complex multi-step reasoning, math, and agentic planning tasks that benefit from deliberate thinking.
Nov 6, 2025262k context0.5x cost rate
Nova Premier 1.0
Amazon Nova Premier is the most capable model in AWS's original Nova lineup, built for complex reasoning, multimodal understanding, and use as a teacher model for distillation. It targets demanding enterprise workloads requiring the highest accuracy in the Nova family.
Oct 31, 20251M context2.5x cost rate
Sonar Pro Search
Sonar Pro Search is a variant of Perplexity's Sonar Pro tuned for heavier, broader web search and retrieval, pulling in more sources per query. It suits research-heavy tasks where thorough web coverage matters more than raw response speed.
Oct 30, 2025200k context3x cost rate
Voxtral Small 24B 2507
Voxtral Small is Mistral AI's 24B audio-native model for speech understanding, transcription, and voice-driven instruction-following. It is designed for tasks combining spoken audio input with text reasoning, such as voice assistants and audio Q&A.
Oct 30, 202532k context0.1x cost rate
gpt-oss-safeguard-20b
GPT-OSS-Safeguard-20B is an open-weight variant of GPT-OSS-20B fine-tuned for safety classification and content moderation, letting developers apply and customize policy-based filtering against custom safety guidelines.
Oct 29, 2025131k context0.1x cost rate
Nemotron Nano 12B 2 VL (free)
Nemotron Nano 12B V2 VL is NVIDIA's compact vision-language model, combining image understanding with efficient text generation. It is suited for lightweight multimodal tasks such as visual question answering and document or image analysis at low cost.
Oct 28, 2025128k contextFree
MiniMax M2
MiniMax-M2 is MiniMax's model built for coding and agentic workflows, offering strong tool-use, multi-step task execution, and software engineering performance. It is designed for developers building autonomous coding agents and complex task automation.
Oct 23, 2025205k context0.2x cost rate
Qwen3 VL 32B Instruct
Qwen3 VL 32B Instruct is a dense multimodal model in Alibaba's Qwen3 VL family, handling image understanding, OCR, and visual reasoning alongside general chat and coding ability.
Oct 23, 2025131k context0.1x cost rate
Granite 4.0 Micro
Granite 4.0 H-Micro is IBM's small, efficient dense language model using a hybrid architecture, optimized for low-latency, cost-efficient workloads. It significantly outperforms earlier same-size Granite models, making it suitable for lightweight chat and instruction-following tasks.
Oct 20, 2025131k context0.1x cost rate
Claude Haiku 4.5
Claude Haiku 4.5 is Anthropic's fast, affordable small model in the Claude 4 generation, balancing strong instruction-following and coding ability with low latency. It is designed for high-throughput agentic and chat applications where speed and cost efficiency are priorities.
Oct 15, 2025200k context1x cost rate
Claude Haiku 4.5 (batch)
Batch-processing variant of Claude Haiku 4.5, Anthropic's fast and affordable Claude 4 small model. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Oct 15, 2025200k context0.5x cost rate
Qwen3 VL 8B Thinking
The extended-reasoning variant of Qwen3 VL 8B, adding step-by-step chain-of-thought to the small multimodal model for improved accuracy on visual reasoning tasks while staying lightweight.
Oct 14, 2025131k context0.4x cost rate
Qwen3 VL 8B Instruct
Qwen3 VL 8B Instruct is a small, fast multimodal model from Alibaba's Qwen3 VL line, suited for lightweight image understanding and vision-plus-text tasks where speed and cost matter most.
Oct 14, 2025262k context0.1x cost rate
Qwen3 VL 30B A3B Thinking
The extended-reasoning variant of Qwen3 VL 30B-A3B, providing step-by-step reasoning over visual and text inputs for more complex multimodal analysis while remaining cheaper than the largest Qwen3 VL models.
Oct 6, 2025262k context0.4x cost rate
Qwen3 VL 30B A3B Instruct
Qwen3 VL 30B-A3B Instruct is a smaller mixture-of-experts multimodal model from Alibaba's Qwen3 VL family, offering solid image and document understanding at lower inference cost than the 235B version.
Oct 6, 2025262k context0.1x cost rate
GPT-5 Pro
GPT-5 Pro is an extended-reasoning tier of GPT-5 that trades latency and cost for higher accuracy on the hardest problems. It is aimed at difficult research, coding, and analytical tasks that benefit from maximum reasoning effort.
Oct 6, 2025400k context23x cost rate
GPT-5 Pro (batch)
GPT-5 Pro is an extended-reasoning tier of GPT-5 for the hardest research, coding, and analytical tasks. This is the batch variant for cheaper asynchronous processing.
Oct 6, 2025400k context12x cost rate
GLM 4.6
GLM-4.6 is Z.ai's open-source flagship model, an incremental upgrade over GLM-4.5 with improved reasoning, coding, and agentic capabilities. It continues the GLM-4.5 architecture lineage while pushing higher benchmark performance for complex, production-grade tasks.
Sep 30, 2025205k context0.4x cost rate
Claude Sonnet 4.5
Claude Sonnet 4.5 is an upgraded mid-tier model in the Claude 4 line, improving coding, computer use, and long-context reasoning over Sonnet 4 while keeping Sonnet-level pricing. It is well suited for enterprise workflows, agentic coding, and general-purpose assistant tasks.
Sep 29, 20251M context3x cost rate
Claude Sonnet 4.5 (batch)
Batch-processing variant of Claude Sonnet 4.5, Anthropic's upgraded mid-tier Claude 4 model. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Sep 29, 20251M context1.5x cost rate
DeepSeek V3.2 Exp
DeepSeek V3.2-exp is an experimental preview checkpoint of DeepSeek V3.2, used to test architecture or training improvements ahead of the stable V3.2 release. It targets the same general reasoning, coding, and chat use cases as V3.2 with less production-hardening.
Sep 29, 2025164k context0.1x cost rate
Cydonia 24B V4.1
Cydonia 24B v4.1 is a community fine-tune from TheDrummer built on a Mistral Small 24B base, tuned for creative writing and roleplay. It favors expressive, character-driven prose over the more clinical tone of general-purpose assistant models.
Sep 27, 2025131k context0.1x cost rate
Relace Apply 3
Relace Apply 3 is a specialized code-patching model that merges AI-suggested edits directly into existing source files at very high speed (thousands of tokens per second). It is designed to be paired with a larger generation model as a fast, cheap 'instant apply' layer, reducing latency and token cost in AI coding tools.
Sep 26, 2025256k context0.4x cost rate
Qwen3 VL 235B A22B Thinking
The extended-reasoning variant of Qwen3 VL 235B-A22B, Alibaba's large multimodal model, which reasons step by step over visual and textual inputs for complex visual reasoning and analysis tasks.
Sep 23, 2025131k context0.7x cost rate
Qwen3 VL 235B A22B Instruct
Qwen3 VL 235B-A22B Instruct is Alibaba's large-scale multimodal MoE model, combining Qwen3's language capability with vision understanding for image, document, and video tasks. It suits demanding multimodal applications needing near-flagship quality.
Sep 23, 2025262k context0.4x cost rate
Qwen3 Max
Qwen3 Max is Alibaba's flagship proprietary Qwen3 model, aimed at top-tier reasoning, coding, and agentic performance to compete with other frontier models like GPT and Claude. It is the highest-capability option in the Qwen3 lineup.
Sep 23, 2025262k context0.8x cost rate
Qwen3 Coder Plus
Qwen3 Coder Plus is a higher-capability tier of Alibaba's Qwen3 Coder family, designed for more demanding agentic coding and software engineering tasks that need stronger reasoning over large codebases than the flash/base variants.
Sep 23, 20251M context0.7x cost rate
GPT-5 Codex (batch)
GPT-5-Codex is a coding-specialized variant of GPT-5 tuned for agentic software engineering in Codex and Codex-like harnesses, handling long-running tasks such as refactors and debugging. This is the batch variant for cheaper asynchronous processing.
Sep 23, 2025400k context0.9x cost rate
DeepSeek V3.1 Terminus
DeepSeek V3.1 Terminus is a refined checkpoint of DeepSeek V3.1, tuned for improved stability, agentic tool use, and coding performance. It is positioned as a polished general-purpose and agentic workhorse ahead of the V3.2/V4 generations.
Sep 22, 2025164k context0.2x cost rate
Qwen3 Coder Flash
Qwen3 Coder Flash is a faster, lower-cost variant of Alibaba's Qwen3 Coder model, trading some capability for speed. It suits latency-sensitive coding assistants and high-volume code-completion workloads.
Sep 17, 20251M context0.2x cost rate
Qwen3 Next 80B A3B Thinking
The extended-reasoning variant of Qwen3-Next 80B-A3B, Alibaba's efficient next-generation MoE model, producing explicit chain-of-thought for harder reasoning and agentic tasks while retaining the architecture's low inference cost.
Sep 11, 2025262k context0.2x cost rate
Qwen3 Next 80B A3B Instruct
Qwen3-Next 80B-A3B Instruct is Alibaba's next-generation efficient mixture-of-experts model (80B total, 3B active), using a highly sparse architecture for strong general-purpose performance at low inference cost.
Sep 11, 2025262k context0.2x cost rate
Qwen Plus 0728
A dated snapshot of Alibaba's Qwen Plus mid-tier model from July 2025, offering the same general-purpose chat, reasoning, and agentic capabilities as Qwen Plus with a pinned version for reproducibility.
Sep 8, 20251M context0.2x cost rate
Qwen Plus 0728 (thinking)
The extended-reasoning (thinking) mode of the July 2025 Qwen Plus snapshot, which produces explicit chain-of-thought before answering. It is suited for harder reasoning and multi-step problem-solving tasks where the standard mode may be less reliable.
Sep 8, 20251M context0.2x cost rate
Nemotron Nano 9B V2 (free)
Nemotron Nano 9B V2 is a compact, efficient open model from NVIDIA's Nemotron family, designed for fast, low-cost text generation and reasoning. It targets latency-sensitive applications where a smaller footprint is preferred over maximum capability.
Sep 5, 2025128k contextFree
Kimi K2 0905
Kimi K2 0905 is a September 2025 update to Moonshot AI's Kimi K2, refining agentic coding and tool-use performance. It keeps the same open-weight, trillion-parameter mixture-of-experts design tuned for long-horizon agent tasks.
Sep 4, 2025262k context0.5x cost rate
Qwen3 30B A3B Thinking 2507
The extended-reasoning (thinking) variant of the July 2025 Qwen3 30B-A3B release, producing explicit chain-of-thought for better performance on reasoning and multi-step tasks while remaining cheap to run.
Aug 28, 202582k context0.4x cost rate
Hermes 4 70B
Hermes 4 70B is Nous Research's fine-tune built on a Llama 3.1 70B base, bringing Hermes 4's improved reasoning, coding, and steerability to a smaller, cheaper, faster model. It suits general chat, roleplay, and agentic tasks with reduced alignment restrictions.
Aug 26, 2025131k context0.1x cost rate
Hermes 4 405B
Hermes 4 405B is Nous Research's next-generation fine-tune built on a Llama 3.1 405B base, improving on Hermes 3 with better reasoning, coding, and structured output while keeping a highly steerable, less-restricted persona. It suits users who want a large open-weight model with strong general capability and flexible alignment.
Aug 26, 2025131k context0.7x cost rate
DeepSeek V3.1
DeepSeek V3.1 is an updated release of DeepSeek's V3 chat model, improving reasoning, coding, and instruction-following over earlier V3 checkpoints. It remains a low-cost, high-throughput option for general-purpose tasks.
Aug 21, 2025164k context0.2x cost rate
Mistral Medium 3.1
Mistral Medium 3.1 is a frontier-class multimodal model from Mistral AI positioned between smaller open models and top proprietary LLMs, supporting long-context reasoning and multi-image input. It has since been superseded by Mistral Medium 3.5, though it remains suitable for general reasoning and vision tasks.
Aug 13, 2025131k context0.4x cost rate
GLM 4.5V
GLM-4.5V is Z.ai's vision-language model built on the GLM-4.5-Air architecture (106B total, 12B active), designed for versatile multimodal reasoning. It handles complex STEM problem-solving, GUI agent tasks, and video understanding via a 3D convolutional vision encoder.
Aug 11, 202566k context0.4x cost rate
GPT-5
GPT-5 is OpenAI's flagship model released in 2025, unifying fast responses and deep reasoning in a single system. It is built for advanced coding, agentic workflows, and complex multi-step problem solving with strong instruction following.
Aug 7, 2025400k context2x cost rate
GPT-5 (batch)
GPT-5 is OpenAI's flagship model unifying fast responses and deep reasoning, built for advanced coding, agentic workflows, and complex problem solving. This is the batch variant for cheaper asynchronous processing.
Aug 7, 2025400k context0.9x cost rate
GPT-5 Mini
GPT-5 Mini is a smaller, faster, and cheaper version of GPT-5 from OpenAI. It targets high-volume chat, lightweight coding, and agentic tasks where speed and cost matter more than maximum reasoning depth.
Aug 7, 2025400k context0.4x cost rate
GPT-5 Mini (batch)
GPT-5 Mini is a smaller, faster, and cheaper version of GPT-5 from OpenAI, suited to high-volume chat, lightweight coding, and agentic tasks. This is the batch variant for cheaper asynchronous processing.
Aug 7, 2025400k context0.2x cost rate
GPT-5 Nano
GPT-5 Nano is the smallest and cheapest model in the GPT-5 family, optimized for extremely fast, low-latency responses. It suits simple classification, extraction, and high-throughput tasks where cost and speed dominate.
Aug 7, 2025400k context0.1x cost rate
GPT-5 Nano (batch)
GPT-5 Nano is the smallest and cheapest model in the GPT-5 family, optimized for fast, low-latency, high-throughput tasks. This is the batch variant for cheaper asynchronous processing.
Aug 7, 2025400k context0.1x cost rate
gpt-oss-120b
GPT-OSS-120B is OpenAI's open-weight mixture-of-experts model (around 120B parameters) released under a permissive license for self-hosted deployment. It offers strong reasoning and coding performance for teams that need to run OpenAI-quality models on their own infrastructure.
Aug 5, 2025131k context0.1x cost rate
gpt-oss-20b
GPT-OSS-20B is OpenAI's smaller open-weight model (around 20B parameters), designed to run on more modest hardware while retaining solid reasoning and coding ability for local or self-hosted use cases.
Aug 5, 2025131k context0.1x cost rate
gpt-oss-20b (free)
GPT-OSS-20B is OpenAI's smaller open-weight model (around 20B parameters), designed to run on modest hardware for local or self-hosted reasoning and coding tasks. This is a free-tier variant of the model.
Aug 5, 2025131k contextFree
Claude Opus 4.1
Claude Opus 4.1 is an incremental upgrade to Claude Opus 4, improving coding accuracy, reasoning, and agentic task performance. It remains Anthropic's top-tier model for complex, high-stakes engineering and analysis work in that generation.
Aug 5, 2025200k context15x cost rate
Claude Opus 4.1 (batch)
Batch-processing variant of Claude Opus 4.1, Anthropic's top-tier model for complex coding and reasoning. Same capabilities as the standard model, offered at lower cost via asynchronous batch processing.
Aug 5, 2025200k context7.5x cost rate
Codestral 2508
Codestral is Mistral AI's dedicated coding model, trained on 80+ programming languages for code generation, completion, and refactoring. It is optimized for low-latency code assistant workflows including fill-in-the-middle completion.
Aug 1, 2025256k context0.2x cost rate
Qwen3 Coder 30B A3B Instruct
A 30B-A3B mixture-of-experts version of Qwen3 Coder, Alibaba's agentic coding model, offering strong code generation and repository-scale understanding at lower inference cost than the full-size Coder model.
Jul 31, 2025262k context0.1x cost rate
Qwen3 30B A3B Instruct 2507
A July 2025 instruction-tuned refresh of Qwen3 30B-A3B, Alibaba's efficient MoE model, with improved instruction following for general chat and task completion at low cost.
Jul 29, 2025262k context0.1x cost rate
GLM 4.5
GLM-4.5 is Z.ai's (Zhipu AI) open-source flagship MoE model (355B total, 32B active) designed for complex reasoning, coding, and agentic workflows. It serves as the base architecture for the broader GLM-4.5 family, including the smaller Air and vision-enabled V variants.
Jul 25, 2025131k context0.5x cost rate
GLM 4.5 Air
GLM-4.5-Air is a lighter-weight, more efficient variant of Z.ai's GLM-4.5 model, trading some capacity for lower cost and faster inference. It is well suited for general reasoning and coding tasks where a smaller footprint is preferred over the full GLM-4.5 model.
Jul 25, 2025131k context0.2x cost rate
Qwen3 235B A22B Thinking 2507
The extended-reasoning variant of the July 2025 Qwen3 235B-A22B release, which generates explicit chain-of-thought for harder reasoning, math, and multi-step agentic tasks before producing a final answer.
Jul 25, 2025262k context0.4x cost rate
Qwen3 Coder 480B A35B
Qwen3 Coder is Alibaba's agentic coding model built on the Qwen3 architecture, tuned for large-context, repository-level code understanding, generation, and tool use. It targets coding agents and IDE assistants that need to reason across whole codebases.
Jul 23, 2025262k context0.2x cost rate
UI-TARS 7B
UI-TARS-1.5-7B is ByteDance Seed's multimodal vision-language agent model built for GUI interaction across desktop, web, and mobile interfaces. It achieves strong results on computer-use benchmarks like OSWorld and AndroidWorld, making it suited for automated screen-based agent tasks rather than general chat.
Jul 22, 2025128k context0.1x cost rate
Gemini 2.5 Flash Lite
Gemini 2.5 Flash-Lite is Google's most lightweight and cost-effective model in the 2.5 family, optimized for very high-throughput, low-latency tasks such as classification, extraction, and simple chat. It trades some reasoning depth for speed and low cost.
Jul 22, 20251M context0.1x cost rate
Gemini 2.5 Flash Lite (batch)
Gemini 2.5 Flash-Lite is Google's most lightweight and cost-effective model in the 2.5 family, optimized for very high-throughput, low-latency tasks such as classification, extraction, and simple chat. This is the batch variant for asynchronous, lower-cost processing.
Jul 22, 20251M context0.1x cost rate
Qwen3 235B A22B Instruct 2507
A July 2025 refreshed release of Qwen3 235B-A22B, Alibaba's large MoE model, with improved instruction following and general capability over the original version. Suited for demanding general-purpose, coding, and agentic workloads.
Jul 21, 2025262k context0.1x cost rate
Kimi K2 0711
Kimi K2 is Moonshot AI's open-weight trillion-parameter mixture-of-experts model built for agentic tasks, coding, and tool use. It is designed as a cost-efficient alternative to top proprietary models for long-horizon agent workflows.
Jul 11, 2025131k context0.5x cost rate
Uncensored
Dolphin Mistral 24B Venice Edition is an uncensored community fine-tune of Mistral 24B, created by Cognitive Computations in collaboration with Venice.ai. It targets general conversation, coding assistance, and creative writing without the guardrails of typical instruction-tuned models.
Jul 9, 2025128k context0.2x cost rate
Hunyuan A13B Instruct
Hunyuan-A13B-Instruct is Tencent's open-source fine-grained MoE model (80B total, 13B active) that adapts its reasoning depth in real time, using a fast path for simple queries and deeper multistep reasoning for complex ones. It is tuned for math, science, and agent-based tasks with a 256k context window.
Jul 8, 2025131k context0.1x cost rate
Morph V3 Large
Morph V3 Large is Morph's high-accuracy code-merging "apply" model, achieving around 98% accuracy on complex edits at moderate speed. It is best used when merge correctness matters more than raw speed, such as applying intricate multi-file diffs from a coding agent.
Jul 7, 2025262k context0.5x cost rate
Morph V3 Fast
Morph V3 Fast is a specialized "apply" model from Morph that merges AI-generated code edits into source files at high speed with around 96% accuracy. It is designed to pair with a coding agent to quickly and reliably apply proposed code diffs rather than generate code itself.
Jul 7, 202582k context0.3x cost rate
ERNIE 4.5 VL 424B A47B
ERNIE 4.5 VL 424B A47B is Baidu's multimodal mixture-of-experts model, with 424B total and 47B active parameters per token, jointly trained on text and image data. It supports both thinking and non-thinking modes for vision-language reasoning, image understanding, and long-context generation up to 131K tokens in English and Chinese.
Jun 30, 2025123k context0.3x cost rate
Mistral Small 3.2 24B
Mistral Small 3.2 is an incremental update to Mistral Small 3.1 from Mistral AI, improving instruction-following and reducing repetition errors. It remains a 24B multimodal model suited for efficient chat, coding, and vision tasks.
Jun 20, 2025256k context0.1x cost rate
MiniMax M1
MiniMax-M1 is MiniMax's large language model built for reasoning, coding, and agentic tool use with efficient long-context processing. It targets developers needing strong reasoning performance at competitive cost.
Jun 17, 20251M context0.5x cost rate
Gemini 2.5 Flash
Gemini 2.5 Flash is Google's fast, cost-efficient multimodal model built for high-volume everyday tasks like chat, summarization, and lightweight agentic workflows. It balances speed and quality with configurable thinking for tasks that need a bit more reasoning.
Jun 17, 20251M context0.5x cost rate
Gemini 2.5 Flash (batch)
Gemini 2.5 Flash is Google's fast, cost-efficient multimodal model built for high-volume everyday tasks like chat, summarization, and lightweight agentic workflows. This is the batch variant for asynchronous, lower-cost processing.
Jun 17, 20251M context0.2x cost rate
Gemini 2.5 Pro
Gemini 2.5 Pro is Google's flagship reasoning model of the 2.5 generation, designed for complex multimodal understanding, advanced coding, and long-context analysis (up to 1M tokens). It excels at multi-step problem solving and agentic tasks that require deep reasoning.
Jun 17, 20251M context2x cost rate
Gemini 2.5 Pro (batch)
Gemini 2.5 Pro is Google's flagship reasoning model of the 2.5 generation, designed for complex multimodal understanding, advanced coding, and long-context analysis. This is the batch variant for asynchronous, lower-cost processing.
Jun 17, 20251M context0.9x cost rate
o3 Pro
o3-pro is a higher-compute version of OpenAI's o3 reasoning model, applying extra thinking effort for maximum reliability on the hardest math, science, and coding problems.
Jun 10, 2025200k context17x cost rate
o3 Pro (batch)
o3-pro is a higher-compute version of OpenAI's o3 reasoning model for the hardest math, science, and coding problems. This is the batch variant for cheaper asynchronous processing.
Jun 10, 2025200k context8x cost rate
Gemini 2.5 Pro Preview 06-05
Gemini 2.5 Pro Preview is the preview release of Google's flagship 2.5 Pro reasoning model, designed for complex multimodal understanding, advanced coding, and long-context analysis. It offers the same strong multi-step reasoning and agentic capabilities as the GA release.
Jun 5, 20251M context2x cost rate
R1 0528
DeepSeek R1 (0528) is an updated checkpoint of DeepSeek's R1 reasoning model with improved chain-of-thought quality and benchmark performance. It targets the same complex math, coding, and logic reasoning use cases as the original R1.
May 28, 2025164k context0.4x cost rate
Claude Opus 4
Claude Opus 4 is Anthropic's flagship model from the Claude 4 generation, built for advanced coding, complex agentic tasks, and deep reasoning. It offers Anthropic's highest intelligence tier for demanding professional and engineering workloads at that generation.
May 22, 2025200k context15x cost rate
Claude Sonnet 4
Claude Sonnet 4 is Anthropic's balanced mid-tier model from the Claude 4 generation, offering strong coding, reasoning, and agentic performance at lower cost than Opus. It is designed as a versatile default for everyday software engineering and assistant tasks.
May 22, 20251M context3x cost rate
Gemma 3n 4B
Gemma 3n E4B IT is Google's open-weight model optimized for on-device and low-memory deployment, using an effective ~4B parameter footprint. It is designed for fast, efficient multimodal (text, image, audio) tasks on phones and edge devices.
May 20, 202533k context0.1x cost rate
Mistral Medium 3
Mistral Medium 3 is a mid-sized frontier model from Mistral AI balancing strong reasoning, coding, and multimodal performance against lower cost than Mistral Large. It targets enterprise and agentic use cases needing near-flagship quality at reduced price.
May 7, 2025131k context0.4x cost rate
Gemini 2.5 Pro Preview 05-06
Gemini 2.5 Pro Preview (05-06) is a dated snapshot of Google's flagship 2.5 Pro reasoning model, designed for complex multimodal understanding, advanced coding, and long-context analysis. It offers the same strong multi-step reasoning and agentic capabilities as later 2.5 Pro releases.
May 7, 20251M context2x cost rate
Virtuoso Large
Virtuoso-Large is Arcee AI's flagship 72B general-purpose model, tuned for cross-domain reasoning, creative writing, and enterprise Q&A with a 128K context window inherited from its Qwen 2.5 base. It combines DeepSeek R1 distillation with SFT and RLHF alignment for strong performance on reasoning and long-context benchmarks.
May 5, 2025131k context0.3x cost rate
Llama Guard 4 12B
Llama Guard 4 12B is Meta's open-weight safety classifier model used to moderate and filter both prompts and model outputs across text and image content. It is designed to be deployed alongside other Llama models to detect unsafe or policy-violating content rather than for general chat.
Apr 30, 20251M context0.1x cost rate
Qwen3 30B A3B
Qwen3 30B-A3B is a smaller mixture-of-experts model (30B total, 3B active parameters) in Alibaba's Qwen3 line, offering good general-purpose capability at low inference cost thanks to its sparse activation.
Apr 28, 2025131k context0.1x cost rate
Qwen3 8B
Qwen3 8B is a small, fast open-weight model in Alibaba's Qwen3 family, suited for lightweight chat, simple coding, and general tasks where speed and low cost matter more than top-end capability.
Apr 28, 2025131k context0.1x cost rate
Qwen3 14B
Qwen3 14B is a mid-size open-weight model in Alibaba's Qwen3 generation, supporting hybrid reasoning and non-reasoning modes. It is designed for general-purpose chat, coding, and agentic tasks at a moderate compute cost.
Apr 28, 2025131k context0.1x cost rate
Qwen3 32B
Qwen3 32B is a dense open-weight model from Alibaba's Qwen3 generation, supporting both reasoning and non-reasoning modes. It is a solid general-purpose choice for chat, coding, and agentic tasks where a mid-size dense model is preferred over MoE.
Apr 28, 2025131k context0.1x cost rate
Qwen3 235B A22B
Qwen3 235B-A22B is a large mixture-of-experts model (235B total, 22B active parameters) from Alibaba's Qwen3 family, offering strong multilingual reasoning, coding, and agentic capabilities. Its MoE design gives near-flagship quality at lower inference cost than an equivalent dense model.
Apr 28, 2025131k context0.4x cost rate
o4 Mini High
o4-mini-high is OpenAI's o4-mini reasoning model run at a higher reasoning-effort setting, improving accuracy on math, coding, and visual reasoning tasks at the cost of some speed.
Apr 16, 2025200k context0.9x cost rate
o4 Mini High (batch)
o4-mini-high is OpenAI's o4-mini reasoning model at a higher reasoning-effort setting for improved accuracy on math, coding, and visual reasoning tasks. This is the batch variant for cheaper asynchronous processing.
Apr 16, 2025200k context0.5x cost rate
o3
OpenAI o3 is a next-generation reasoning model succeeding o1, with stronger performance on math, science, coding, and multi-step logical problem solving through deeper internal deliberation.
Apr 16, 2025200k context1.5x cost rate
o3 (batch)
OpenAI o3 is a reasoning model with strong math, science, coding, and multi-step logical problem-solving performance. This is the batch variant for cheaper asynchronous processing.
Apr 16, 2025200k context0.8x cost rate
o4 Mini
OpenAI o4-mini is an efficient reasoning model succeeding o3-mini, combining strong math, coding, and visual reasoning performance with low cost and fast response times.
Apr 16, 2025200k context0.9x cost rate
o4 Mini (batch)
OpenAI o4-mini is an efficient reasoning model with strong math, coding, and visual reasoning performance at low cost. This is the batch variant for cheaper asynchronous processing.
Apr 16, 2025200k context0.5x cost rate
GPT-4.1
GPT-4.1 is OpenAI's 2025 flagship model, improving coding, instruction-following, and long-context handling (up to 1M tokens) over GPT-4o. It is well suited for agentic coding, complex reasoning, and tasks requiring very long context.
Apr 14, 20251M context1.5x cost rate
GPT-4.1 (batch)
GPT-4.1 is OpenAI's 2025 flagship model, improving coding, instruction-following, and long-context handling (up to 1M tokens) over GPT-4o, well suited for agentic coding and complex reasoning. This is the batch variant for asynchronous, lower-cost processing.
Apr 14, 20251M context0.8x cost rate
GPT-4.1 Mini
GPT-4.1 Mini is a smaller, faster, cheaper variant of GPT-4.1 that retains strong coding and reasoning ability with a large context window. It is suited for high-volume or latency-sensitive applications that still need solid general capability.
Apr 14, 20251M context0.3x cost rate
GPT-4.1 Mini (batch)
GPT-4.1 Mini is a smaller, faster, cheaper variant of GPT-4.1 that retains strong coding and reasoning ability with a large context window, suited for high-volume or latency-sensitive applications. This is the batch variant for asynchronous, lower-cost processing.
Apr 14, 20251M context0.2x cost rate
GPT-4.1 Nano
GPT-4.1 Nano is OpenAI's smallest and fastest GPT-4.1 variant, optimized for extremely low-latency, low-cost tasks such as classification, extraction, and simple chat. It sacrifices some capability for speed and price relative to Mini and full GPT-4.1.
Apr 14, 20251M context0.1x cost rate
GPT-4.1 Nano (batch)
GPT-4.1 Nano is OpenAI's smallest and fastest GPT-4.1 variant, optimized for extremely low-latency, low-cost tasks such as classification, extraction, and simple chat. This is the batch variant for asynchronous, lower-cost processing.
Apr 14, 20251M context0.1x cost rate
Llama 4 Maverick
Llama 4 Maverick is Meta's high-capability open-weight Mixture-of-Experts model, offering strong multimodal (text and image) understanding, reasoning, and coding at efficient inference cost. It targets general-purpose assistant and agentic use cases needing frontier-adjacent quality in an open model.
Apr 5, 20251M context0.2x cost rate
Llama 4 Scout
Llama 4 Scout is Meta's efficient open-weight Mixture-of-Experts model with an industry-leading long context window, designed for tasks requiring extensive context handling alongside multimodal understanding. It offers a lighter, faster alternative to Llama 4 Maverick.
Apr 5, 20251.3M context0.1x cost rate
DeepSeek V3 0324
DeepSeek V3 (0324) is a checkpoint of DeepSeek's V3 mixture-of-experts model, offering strong general reasoning, coding, and instruction-following. It is a cost-efficient open-weight alternative to closed frontier models.
Mar 24, 2025164k context0.2x cost rate
o1-pro
o1-pro is a higher-compute version of OpenAI's o1 reasoning model, applying more thinking effort for greater reliability on the hardest math, science, and coding problems.
Mar 19, 2025200k context125x cost rate
o1-pro (batch)
o1-pro is a higher-compute version of OpenAI's o1 reasoning model for the hardest math, science, and coding problems. This is the batch variant for cheaper asynchronous processing.
Mar 19, 2025200k context63x cost rate
Mistral Small 3.1 24B
Mistral Small 3.1 is a 24B open-weight multimodal model from Mistral AI supporting vision input alongside text, with a large context window. It is built for fast, cost-effective chat, coding, and image-understanding tasks.
Mar 17, 2025128k context0.2x cost rate
Gemma 3 4B
Gemma 3 4B IT is a small open-weight instruction-tuned model from Google's Gemma 3 family, designed for fast, low-cost inference on edge or resource-constrained hardware. It handles basic chat, summarization, and simple multimodal tasks efficiently.
Mar 13, 2025131k context0.1x cost rate
Gemma 3 12B
Gemma 3 12B IT is a mid-size open-weight instruction-tuned model from Google's Gemma 3 family, supporting multimodal (text and image) input and long context. It is well suited for general-purpose chat, reasoning, and coding assistance on self-hosted or resource-constrained deployments.
Mar 13, 2025131k context0.1x cost rate
Command A
Command A is Cohere's flagship enterprise model, optimized for agentic tool use, RAG, multilingual business tasks, and instruction-following at efficient serving cost. It is designed for enterprise deployments needing strong accuracy with lower compute requirements than typical frontier models.
Mar 13, 2025256k context2x cost rate
Reka Flash 3
Reka Flash 3 is a mid-size multimodal model from Reka AI balancing speed and capability, supporting text and vision inputs for general chat, reasoning, and multimodal tasks at a favorable cost-performance ratio.
Mar 12, 202566k context0.1x cost rate
Gemma 3 27B
Gemma 3 27B IT is the largest dense model in Google's open-weight Gemma 3 family, supporting multimodal input and strong multilingual capability. It targets general chat, reasoning, and coding tasks where higher quality than smaller Gemma variants is needed while staying open and self-hostable.
Mar 12, 2025262k context0.1x cost rate
Skyfall 36B V2
Skyfall 36B v2 is a larger creative-writing and roleplay fine-tune from TheDrummer, offering more capacity and nuance than the smaller Rocinante and Cydonia models in the same lineup. It is designed for expressive, character-consistent storytelling.
Mar 10, 202533k context0.2x cost rate
Sonar Reasoning Pro
Sonar Reasoning Pro combines Perplexity's web-grounded search with explicit chain-of-thought reasoning. It is designed for complex analytical questions that require both up-to-date factual grounding and multi-step logical reasoning over the retrieved information.
Mar 7, 2025128k context1.5x cost rate
Sonar Pro
Sonar Pro is Perplexity's higher-quality, search-grounded model with a larger context window and more thorough citation handling than the base Sonar. It is designed for complex, multi-part queries that need accurate, well-sourced, up-to-date answers.
Mar 7, 2025200k context3x cost rate
Sonar Deep Research
Sonar Deep Research is Perplexity's autonomous research model that performs multi-step web searches, reads many sources, and synthesizes long, well-cited reports. It is suited for comprehensive research tasks, literature-style summaries, and in-depth fact-finding that goes far beyond a single search query.
Mar 7, 2025128k context1.5x cost rate
Saba
Mistral Saba is a regional model from Mistral AI tailored for Arabic and languages of the Middle East and South Asia. It is designed for culturally and linguistically accurate text generation and chat in those regions.
Feb 17, 202533k context0.1x cost rate
o3 Mini High
o3-mini-high is OpenAI's o3-mini reasoning model run at a higher reasoning-effort setting, trading some speed for improved accuracy on math, coding, and logic tasks.
Feb 12, 2025200k context0.9x cost rate
o3 Mini High (batch)
o3-mini-high is OpenAI's o3-mini reasoning model at a higher reasoning-effort setting for improved accuracy on math, coding, and logic tasks. This is the batch variant for cheaper asynchronous processing.
Feb 12, 2025200k context0.5x cost rate
Aion-RP 1.0 (8B)
Aion-RP-Llama-3.1-8B is a lightweight roleplay-focused fine-tune of Meta's Llama 3.1 8B from AionLabs. It is designed for fast, low-cost character roleplay and creative fiction rather than complex reasoning or coding.
Feb 4, 202533k context0.4x cost rate
Qwen2.5 VL 72B Instruct
Qwen 2.5 VL 72B Instruct is Alibaba's large open-weight vision-language model, capable of image understanding, document/OCR parsing, and visual reasoning alongside text. It is designed for multimodal tasks that combine strong language ability with detailed visual comprehension.
Feb 1, 2025128k context0.3x cost rate
Qwen-Plus
Qwen Plus is Alibaba's proprietary mid-tier model in the Qwen lineup, balancing cost and capability for general chat, reasoning, and agentic tasks. It sits between the lightweight and flagship Qwen offerings for everyday production use.
Feb 1, 20251M context0.2x cost rate
o3 Mini
o3-mini is a smaller, faster, and cheaper version of OpenAI's o3 reasoning model, aimed at math, coding, and logic tasks where lower latency and cost matter more than maximum reasoning depth.
Jan 31, 2025200k context0.9x cost rate
o3 Mini (batch)
o3-mini is a smaller, faster, and cheaper version of OpenAI's o3 reasoning model for math, coding, and logic tasks. This is the batch variant for cheaper asynchronous processing.
Jan 31, 2025200k context0.5x cost rate
Mistral Small 3
Mistral Small 3 (24B, January 2025) is a dense open-weight model from Mistral AI tuned for low-latency chat, coding, and instruction-following. It is designed to deliver strong performance at a fraction of the cost of larger models.
Jan 30, 202533k context0.1x cost rate
Sonar
Sonar is Perplexity's fast, search-augmented language model that grounds answers in real-time web results with citations. It is designed for quick, factual question answering and up-to-date information retrieval rather than deep reasoning or creative tasks.
Jan 27, 2025127k context0.3x cost rate
R1 Distill Llama 70B
DeepSeek R1 Distill Llama 70B is a Llama 3.3 70B model distilled from DeepSeek R1's reasoning traces, transferring much of R1's chain-of-thought reasoning ability into a smaller, faster base architecture. It suits math, coding, and logic tasks where full R1 would be too costly or slow.
Jan 23, 20258k context0.3x cost rate
R1
DeepSeek R1 is DeepSeek's dedicated reasoning model, trained with reinforcement learning to produce explicit chain-of-thought before answering. It is designed for complex math, coding, and logic problems, and was notable for matching top closed reasoning models at a fraction of the cost.
Jan 20, 202564k context0.5x cost rate
MiniMax-01
MiniMax-01 is MiniMax's foundational large language model, an early entry in the MiniMax series offering general-purpose chat, reasoning, and long-context handling. It established the base architecture later refined into the MiniMax M-series.
Jan 15, 20251M context0.2x cost rate
Phi 4
Phi-4 is Microsoft's small open-weight language model that punches above its parameter count, particularly strong on reasoning and math tasks relative to its size. It is well suited for efficient, cost-conscious deployments needing solid reasoning without a large model footprint.
Jan 10, 202516k context0.1x cost rate
DeepSeek V3
DeepSeek Chat is DeepSeek's general-purpose conversational model, corresponding to the non-thinking mode of the DeepSeek V3 model family. It is designed for everyday chat, coding assistance, and reasoning tasks at low cost.
Dec 26, 2024164k context0.2x cost rate
Llama 3.3 Euryale 70B
L3.3 Euryale 70B is a Llama 3.3 70B fine-tune by Sao10K, continuing the Euryale line's focus on immersive roleplay and creative writing with improved base-model capability.
Dec 18, 2024131k context0.2x cost rate
o1
OpenAI o1 is the first model in OpenAI's dedicated reasoning line, using extended internal chain-of-thought before answering. It is designed for hard math, science, and coding problems that benefit from deliberate step-by-step reasoning.
Dec 17, 2024200k context13x cost rate
o1 (batch)
OpenAI o1 is the first model in OpenAI's reasoning line, using extended chain-of-thought for hard math, science, and coding problems. This is the batch variant for cheaper asynchronous processing.
Dec 17, 2024200k context6x cost rate
Command R7B (12-2024)
Command R7B (12-2024) is Cohere's compact model in the Command R family, designed for fast, low-cost RAG, tool use, and conversational tasks on resource-constrained deployments. It trades some capability for speed and efficiency versus the larger Command R models.
Dec 14, 2024128k context0.1x cost rate
Llama 3.3 70B Instruct
Llama 3.3 70B Instruct is Meta's open-weight model that delivers performance close to Llama 3.1 405B at a fraction of the size and cost, with a 128K context window. It is well suited for general chat, reasoning, and coding in self-hosted deployments.
Dec 6, 2024131k context0.1x cost rate
Nova Lite 1.0
Amazon Nova Lite is a low-cost multimodal model from AWS's original Nova family, handling text, image, and video input with fast response times. It is designed for high-volume, latency-sensitive tasks like summarization and simple assistant workflows.
Dec 5, 2024300k context0.1x cost rate
Nova Micro 1.0
Amazon Nova Micro is a text-only, ultra-low-latency model from AWS's Nova family, optimized purely for speed and cost efficiency. It suits simple, high-throughput text tasks like classification, extraction, and short-form generation.
Dec 5, 2024128k context0.1x cost rate
Nova Pro 1.0
Amazon Nova Pro is a balanced multimodal model from AWS's Nova family, handling text, image, and video input with strong accuracy at moderate cost and latency. It is designed as an all-round workhorse for agentic and enterprise applications.
Dec 5, 2024300k context0.7x cost rate
GPT-4o (2024-11-20)
GPT-4o 2024-11-20 is a pinned November 2024 snapshot of OpenAI's GPT-4o with improved creative writing quality and instruction-following. It retains the same fast multimodal capabilities as GPT-4o for general-purpose chat and vision tasks.
Nov 20, 2024128k context2x cost rate
Mistral Large 2407
Mistral Large 2 (2407) is Mistral AI's mid-2024 flagship model with strong multilingual, coding, and reasoning capabilities and a 128K context window. It was built as a high-capability alternative to top proprietary models for enterprise use cases.
Nov 19, 2024131k context1x cost rate
Qwen2.5 Coder 32B Instruct
Qwen 2.5 Coder 32B Instruct is Alibaba's dedicated code-generation model, specialized for writing, completing, and explaining code across many programming languages. It is well suited for coding assistants and developer tooling that need strong code-specific performance.
Nov 11, 202433k context0.3x cost rate
UnslopNemo 12B
UnslopNemo 12B is a TheDrummer fine-tune of Mistral Nemo 12B aimed at reducing repetitive, generic "AI-isms" (slop) in generated prose. It is built for creative writing and roleplay where more natural, varied language is desired.
Nov 8, 20241M context0.1x cost rate
Magnum v4 72B
Magnum v4 72B is a community fine-tune from Anthracite based on Qwen2.5 72B, trained to emulate the prose quality and creative writing style of Claude 3 models. It is aimed at long-form creative writing and roleplay rather than reasoning or coding tasks.
Oct 22, 202433k context1x cost rate
Ministral 8B
Ministral 8B is Mistral AI's October 2024 edge-focused dense model (8B parameters), using an interleaved sliding-window attention pattern for fast, memory-efficient inference and supporting up to 128K context, available under Mistral's research and commercial licenses. It targets on-device and privacy-critical deployments such as local translation, offline assistants, and robotics, and doubles as a low-latency function-calling intermediary for multi-step agentic workflows; it was superseded in this role by the December 2025 Ministral 3 8B.
Oct 17, 2024128k context0.1x cost rate
Qwen2.5 7B Instruct
Qwen 2.5 7B Instruct is a compact open-weight chat model from Alibaba's Qwen family, tuned for fast, low-cost general-purpose tasks. It trades some capability for speed and efficiency compared to the larger Qwen 2.5 models.
Oct 16, 202433k context0.1x cost rate
Rocinante 12B
Rocinante 12B is a TheDrummer fine-tune of a 12B Mistral Nemo base model, optimized for creative writing and roleplay use cases. It is a lightweight option for character-driven storytelling rather than general assistant tasks.
Sep 30, 202466k context0.1x cost rate
Llama 3.2 1B Instruct
Llama 3.2 1B Instruct is Meta's smallest open-weight instruction-tuned model, built for extremely low-latency, on-device or edge deployment. It handles basic text tasks like classification and simple chat rather than complex reasoning.
Sep 25, 202460k context0.1x cost rate
Llama 3.2 3B Instruct
Llama 3.2 3B Instruct is a small open-weight instruction-tuned model from Meta, suited for on-device and edge deployment where efficiency matters. It handles lightweight chat, summarization, and simple instruction-following tasks.
Sep 25, 2024131k context0.1x cost rate
Qwen2.5 72B Instruct
Qwen 2.5 72B Instruct is Alibaba's large general-purpose open-weight chat model, strong at multilingual conversation, reasoning, and coding. It suits demanding general assistant tasks where a bigger dense model improves quality over smaller Qwen variants.
Sep 19, 202433k context0.1x cost rate
Command R (08-2024)
Command R (08-2024) is a mid-sized model from Cohere's Command R family, optimized for retrieval-augmented generation, tool use, and multilingual conversational tasks. It balances capability and cost for enterprise RAG and chat applications.
Aug 30, 2024128k context0.1x cost rate
Command R+ (08-2024)
Command R+ (08-2024) is Cohere's larger, more capable model in the Command R family, built for complex RAG pipelines, multi-step tool use, and enterprise-grade multilingual tasks. It offers higher accuracy than Command R at increased cost.
Aug 30, 2024128k context2x cost rate
Llama 3.1 Euryale 70B v2.2
L3.1 Euryale 70B is a Llama 3.1 70B fine-tune by Sao10K, optimized for roleplay, creative writing, and character-driven storytelling with rich, expressive prose.
Aug 28, 2024131k context0.3x cost rate
Hermes 3 70B Instruct
Hermes 3 70B is Nous Research's fine-tune of Meta's Llama 3.1 70B, offering the same steerable, low-refusal instruction-following and roleplay focus as the 405B version in a smaller, faster package. It suits general chat, creative writing, and agentic use where lower cost matters.
Aug 18, 2024131k context0.2x cost rate
Hermes 3 405B Instruct
Hermes 3 405B is Nous Research's fine-tune of Meta's Llama 3.1 405B, tuned for strong instruction-following, roleplay, and agentic reasoning with reduced refusals. It suits users who want a large, highly steerable open-weight model with fewer alignment restrictions than the base model.
Aug 16, 2024131k context0.3x cost rate
Llama 3 8B Lunaris
L3 Lunaris 8B is a community fine-tune of Meta's Llama 3 8B by Sao10K, tuned for roleplay and creative writing. It is designed for immersive character-driven storytelling rather than general assistant tasks.
Aug 13, 20248k context0.1x cost rate
GPT-4o (2024-08-06)
GPT-4o 2024-08-06 is a pinned August 2024 snapshot of OpenAI's GPT-4o that added support for structured JSON outputs. It retains GPT-4o's fast multimodal text, image, and audio capabilities for general-purpose chat and vision tasks.
Aug 6, 2024128k context2x cost rate
Llama 3.1 70B Instruct
Llama 3.1 70B Instruct is Meta's open-weight instruction-tuned model, offering strong general reasoning, chat, and coding performance with a 128K context window. It is a popular, well-balanced choice for self-hosted deployments needing near-frontier quality at open-weight cost.
Jul 23, 2024131k context0.1x cost rate
Llama 3.1 8B Instruct
Llama 3.1 8B Instruct is Meta's small open-weight instruction-tuned model, designed for fast, low-cost chat and general text tasks with a 128K context window. It suits lightweight applications and resource-constrained self-hosted deployments.
Jul 23, 2024131k context0.1x cost rate
Mistral Nemo
Mistral Nemo is a 12B-parameter multilingual model built by Mistral AI in collaboration with NVIDIA, featuring a 128K context window. It is designed for efficient reasoning, coding, and multilingual chat with a strong performance-to-size ratio.
Jul 19, 2024131k context0.1x cost rate
GPT-4o-mini
GPT-4o Mini is OpenAI's small, low-cost multimodal model, offering solid reasoning and vision capability at a fraction of GPT-4o's price. It is well suited for high-volume, cost-sensitive applications like chatbots and lightweight assistants.
Jul 18, 2024128k context0.1x cost rate
GPT-4o-mini (2024-07-18)
GPT-4o Mini 2024-07-18 is the pinned initial-release snapshot of OpenAI's small, low-cost GPT-4o Mini model from July 2024. It offers the same fast, affordable multimodal reasoning and vision capability as GPT-4o Mini, kept fixed for reproducibility.
Jul 18, 2024128k context0.1x cost rate
GPT-4o-mini (batch)
GPT-4o Mini is OpenAI's small, low-cost multimodal model, offering solid reasoning and vision capability at a fraction of GPT-4o's price, well suited for high-volume, cost-sensitive applications. This is the batch variant for asynchronous, lower-cost processing.
Jul 18, 2024128k context0.1x cost rate
Gemma 2 27B
Gemma 2 27B IT is Google's open-weight instruction-tuned language model from the Gemma 2 family, suitable for general chat, reasoning, and text generation tasks that can be self-hosted. It offers strong quality for its size while remaining runnable on modest hardware.
Jul 13, 20248k context0.2x cost rate
GPT-4o
GPT-4o is OpenAI's natively multimodal flagship model handling text, image, and audio input with fast response times. It is designed for general-purpose chat, vision tasks, and real-time conversational applications.
May 13, 2024128k context2x cost rate
GPT-4o (2024-05-13)
GPT-4o 2024-05-13 is the pinned initial-release snapshot of OpenAI's natively multimodal GPT-4o model from May 2024. It offers the same fast text, image, and audio understanding as GPT-4o, kept fixed for reproducibility.
May 13, 2024128k context3x cost rate
GPT-4o (batch)
GPT-4o is OpenAI's original natively multimodal flagship model, handling text, vision, and audio in one system. It is a solid general-purpose choice for chat, vision tasks, and everyday coding; this is the batch variant for cheaper asynchronous processing.
May 13, 2024128k context1x cost rate
Mixtral 8x22B Instruct
Mixtral 8x22B is Mistral AI's sparse mixture-of-experts model with 8 experts of 22B parameters each and roughly 39B active parameters. It offers strong reasoning, multilingual, and coding performance at efficient inference cost relative to a dense model of similar quality.
Apr 17, 202466k context1x cost rate
WizardLM-2 8x22B
WizardLM-2 8x22B is Microsoft's open-weight Mixture-of-Experts model tuned for complex instruction following, reasoning, and multilingual chat. It offers strong general-purpose capability among open models, suited for advanced conversational and reasoning tasks.
Apr 16, 202466k context0.2x cost rate
GPT-4 Turbo
GPT-4 Turbo is OpenAI's faster, cheaper evolution of GPT-4 with a larger context window and a more recent knowledge cutoff. It is suited for complex reasoning, coding, and long-document tasks at lower cost than the original GPT-4.
Apr 9, 2024128k context6.5x cost rate
GPT-4 Turbo (batch)
GPT-4 Turbo is OpenAI's faster, cheaper evolution of GPT-4 with a larger context window, suited for complex reasoning, coding, and long-document tasks. This is the batch variant for asynchronous, lower-cost processing.
Apr 9, 2024128k context3x cost rate
Claude 3 Haiku
Claude 3 Haiku is Anthropic's fastest and most affordable model from the original Claude 3 generation, designed for near-instant responses on simple tasks. It suits high-volume use cases like customer support, content moderation, and quick Q&A where speed and cost matter more than deep reasoning.
Mar 13, 2024200k context0.3x cost rate
Mistral Large
Mistral Large is Mistral AI's flagship general-purpose model line for complex reasoning, coding, and multilingual tasks, with this key generally pointing at the current default Mistral Large release. It is designed for demanding enterprise workloads requiring strong reasoning and instruction-following.
Feb 26, 2024128k context1x cost rate
GPT-3.5 Turbo (older v0613)
GPT-3.5 Turbo 0613 is a pinned June 2023 snapshot of OpenAI's GPT-3.5 Turbo chat model, kept for reproducibility. It offers the same fast, low-cost chat capability as GPT-3.5 Turbo, suited for simple conversational and summarization tasks.
Jan 25, 20244k context0.5x cost rate
GPT-4 Turbo Preview
GPT-4 Turbo Preview is a preview snapshot of OpenAI's GPT-4 Turbo, offering the same faster, cheaper reasoning and coding capability with a large context window ahead of the stable release. It suits the same complex reasoning and long-document use cases as GPT-4 Turbo.
Jan 25, 2024128k context6.5x cost rate
GPT-3.5 Turbo Instruct
GPT-3.5 Turbo Instruct is OpenAI's completion-style (non-chat) variant of GPT-3.5 Turbo, designed for legacy instruction-following completion APIs rather than multi-turn chat. It is suited for simple text completion and instruction tasks in legacy integrations.
Sep 28, 20234k context0.6x cost rate
GPT-3.5 Turbo 16k
GPT-3.5 Turbo 16K is a variant of OpenAI's GPT-3.5 Turbo with an extended 16K-token context window, otherwise sharing the same fast, low-cost chat capabilities. It suits tasks needing longer input or output than the standard 4K context allows.
Aug 28, 202316k context1x cost rate
Weaver (alpha)
Weaver (alpha) from Mancer is a fine-tune on top of Qwen2.5 72B designed to recreate Claude-style prose verbosity for roleplay and narrative writing. It is aimed at creative and roleplay use cases rather than coherence-critical or agentic tasks, with a modest 8K context window.
Aug 2, 20238k context0.2x cost rate
ReMM SLERP 13B
RemM SLERP L2 13B is a community model merge (using SLERP interpolation) built on Llama 2 13B by well-known model merger Undi95. It is a classic roleplay- and creative-writing-oriented model rather than a general assistant.
Jul 22, 20236k context0.2x cost rate
MythoMax 13B
MythoMax L2 13B is a popular community fine-tune (Gryphe) built on Llama 2 13B, blending several roleplay and storytelling models. It is designed for creative writing, roleplay, and character-driven chat rather than technical or agentic tasks.
Jul 2, 20238k context0.1x cost rate
GPT-3.5 Turbo
GPT-3.5 Turbo is OpenAI's earlier-generation chat model, fast and inexpensive but less capable than GPT-4-class models. It is suited for simple chat, summarization, and lightweight tasks where cost and speed outweigh the need for advanced reasoning.
May 28, 202316k context0.3x cost rate
GPT-3.5 Turbo (batch)
GPT-3.5 Turbo is OpenAI's earlier-generation chat model, fast and inexpensive, suited for simple chat, summarization, and lightweight tasks. This is the batch variant for asynchronous, lower-cost processing.
May 28, 202316k context0.2x cost rate
GPT-4
GPT-4 is OpenAI's landmark large multimodal model offering strong reasoning, coding, and instruction-following well beyond GPT-3.5. It is suited for complex tasks requiring deeper reasoning, though it has largely been superseded by faster, cheaper GPT-4o and GPT-4.1 models.
May 28, 20238k context15x cost rate

Models

DeepSeek V4 Flash Vision Exp

Ox Alpha

Hy-MT2-1.8B

Hy-MT2-30B-A3B

GLM Latest

GLM 5.3

Qwen3.8 27B

Dots3-Note Preview (free)

Gemini 3.7 Flash

Gemini 3.7 Flash (batch)

Seed 2.1 Turbo

Qwen3.8 2.4T A95B

Seed-2.0-Code

DeepSeek V4 Pro 0813

Grok 4.6

LFM2.5-2.6B (free)

Nemotron 3.5 Lightning

Nemotron 3.5 Lightning (free)

Sakana Namazu

Solar Pro 4

Muse Glimmer 30B

Muse Spark 1.2

Qwen3.8 Max

DeepSeek V4 Flash Latest

DeepSeek V4 Flash 0731

Inkling Small

Qwen3.7 Flash

Claude Opus 5 (Fast)

Claude Opus 5

Claude Opus 5 (batch)

Ling-3.0-flash

Laguna S 2.1

Laguna S 2.1 (free)

Gemini 3.6 Flash

Gemini 3.6 Flash (batch)

Gemini 3.5 Flash Lite

Gemini 3.5 Flash Lite (batch)

LongCat 2.0

Inkling

Inkling (batch)

Kimi K3

Muse Spark 1.1

KAT-Coder-Air V2.5

KAT-Coder-Pro V2.5

GPT-5.6 Luna Pro

GPT-5.6 Luna Pro (batch)

GPT-5.6 Luna

GPT-5.6 Luna (batch)

GPT-5.6 Terra Pro

GPT-5.6 Terra Pro (batch)

GPT-5.6 Terra

GPT-5.6 Terra (batch)

GPT-5.6 Sol Pro

GPT-5.6 Sol Pro (batch)

GPT-5.6 Sol

GPT-5.6 Sol (batch)

Grok 4.5

Grok Latest

Aion-3.0-Mini

Aion-3.0

Hy3

Laguna XS 2.1

Laguna XS 2.1 (free)

Claude Sonnet 5

Claude Sonnet 5 (batch)

Nex-N2-Mini

Fugu Ultra

North Mini Code (free)

GLM 5.2

GLM 5.2 (batch)

GLM 5.2 (free)

Kimi K2.7 Code

Kimi K2.7 Code (batch)

Claude Fable Latest

Claude Fable 5

Claude Fable 5 (batch)

Nex-N2-Pro

Nemotron 3.5 Content Safety (free)

Nemotron 3 Ultra