DeepSeek V4.1 Flash
DeepSeek馃嚚馃嚦
deepseek/deepseek-v4.1-flashDeepSeek V4.1 Flash is DeepSeek's open-weight sparse MoE model and the first built on the company's Causal Encoder-Decoder architecture, activating 8B parameters on input and 16B on output with a 1M-token context window. It targets coding, terminal and computer-use agents, and long-horizon multi-step tasks as a low-cost, high-throughput sibling to DeepSeek's Pro line.
Cost rate
0.1x
Context
1M
Released
Sep 10, 2026
Input
TextImage
Output
Text
Support
ReasoningTool callingStructured outputs
Best at
The categories where this model ranks highest.
Medicine & Healthcare
#16 路 top 4%
Software & IT Services
#25 路 top 6%
Mathematical
#28 路 top 7%
Business, Management, & Finance
#30 路 top 7%
Performance
Median latency and throughput measured across recent requests.
Throughput234 tok/s
Latency