DeepSeek
DeepSeek: R1
deepseek/deepseek-r1
DeepSeek R1 is here: Performance on par with [OpenAI o1](/openai/o1), but open-sourced and with fully open reasoning tokens. It's 671B parameters in size, with 37B active in an inference pass.
- Context window: 64,000 tokens
- Input: text
- Output: text
- Pricing: $0.7/M input tokens, $2.5/M output tokens
RuntimeWire coverage
- OpenAI reports up to 3.6x lower Jalapeno latency in engineering tests
- OpenAI says Jalapeno beats Nvidia Blackwell on inference speed per watt
- DeepSeek warns of a significant API price increase after undercutting US rivals
- CrucibleBench finds an LLM judge shifted rankings by six spots
- Sean Du brings a reasoning-model hallucination detector to ICML 2026
- Tensordyne puts log math at the center of its Nvidia challenge
- DeepSeek-R1 beats Codestral-2501 where it counts
View on OpenRouter. Model data sourced from OpenRouter.