deepseek
DeepSeek: DeepSeek V4.1 Flash
deepseek/deepseek-v4.1-flash
DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture.
- Context window: 1,048,576 tokens
- Input: text, image
- Output: text
- Pricing: $0.045/M input tokens, $0.5/M output tokens
RuntimeWire coverage
- Replit's ViBench 2.0 gives top AI models a 49%-65% whole-product pass rate
- Suhail Doshi launches Particle to cut coding agents' token use
- AllModels says request IDs can trace refusal-removed model use back to customers
- One DeepSeek model, 41 endpoints, a 13x price gap on OpenRouter
- DeepSeek opens its inference engine, leaving service plumbing to developers
- Infron offers six AI model endpoints at $0 through one API
- Together AI adds DeepSeek's coding update to its API for agent workloads
- RuntimeWire flagged DeepSeek's V4.1-Flash-1017 refresh 3 hours before its announcement
- Neuralwatt says cache misses can use up to 140% more inference energy
- Two coding-agent tests report DeepSeek V4.1 Flash used exposed API keys
- Head to head: DeepSeek-V4.1-Flash vs Mistral-Large-3
- Head to head: DeepSeek-V4.1-Flash vs muse-spark-1.3
View on OpenRouter. Model data sourced from OpenRouter.