We are an open-source initiative to democratize reinforcement learning (RL) techniques and develop scalable systems for large language models (LLMs) and agents.
We release DeepCoder-14B-Preview, a code reasoning model finetuned from Deepseek-R1-Distilled-Qwen-14B via distributed RL. It achieves an impressive 60.6% Pass@1 accuracy on LiveCodeBench (+8% improvement), matching the performance of o3-mini-2025-01-031 (Low) and o1-2024-12-17 with just 14B parameters.
DeepScaleR is an open-source effort to fully democratize reinforcement learning (RL) for LLMs and reproduce DeepSeek R1 and OpenAI O1/O3 at scale on real tasks. We introduce DeepScaleR-1.5B-Preview, a language model finetuned from Deepseek-R1-Distilled-Qwen-1.5B using distributed reinforcement learning (RL). It achieves an impressive 43.1% Pass@1 accuracy on AIME 2024 surpassing the performance of OpenAI's o1-preview with just 1.5B parameters...
