1
DeepSeek Open Sources DeepSeek R1 LLM with Performance Comparable To OpenAI's O1 Model
linneabridges6 edited this page 2 weeks ago
DeepSeek open-sourced DeepSeek-R1, an LLM fine-tuned with support learning (RL) to improve thinking ability. DeepSeek-R1 attains results on par with OpenAI's o1 design on numerous standards, including MATH-500 and wiki.whenparked.com SWE-bench.
DeepSeek-R1 is based on DeepSeek-V3, a mixture of experts (MoE) design recently open-sourced by DeepSeek. This base model is fine-tuned using Group Relative Policy Optimization (GRPO), a reasoning-oriented variation of RL. The research group likewise carried out understanding distillation from DeepSeek-R1 to open-source Qwen and Llama models and released a number of versions of each