Awesome-LLM-Strawberry

Why we included this project

Reasoning models like OpenAI o1 are hard to follow because the material that explains them is scattered across official docs, lab blog posts, and research papers. This index collects the primary sources in one place: OpenAI's own documentation and announcements, write-ups from labs like DeepSeek and Google DeepMind, and the key papers on chain-of-thought, reinforcement learning, and inference-time scaling. The open-source section is the most practical part, pointing to models and training codebases you can actually inspect or run, from Qwen's QwQ to DeepSeek R1 and RLHF frameworks like OpenRLHF. It is a reading list and research map rather than a deployable tool, and it is updated as the field moves. Teams trying to understand or replicate o1-style training will save real time starting here.