How deepseek-ai/deepseek-r1 works
This repository is a research and model-release landing page for DeepSeek-R1, not an implementation or serving runtime: its tree contains a README, the paper PDF, and figures, with no source-code modules. The documented lifecycle spans base-model RL, cold-start/SFT and RL post-training, reasoning-data distillation into dense models, and evaluation. Links among these stages are stated in the README; no implementation-level calls or data paths can be established from the available files.
Components 11
Model development
- V3 base model
- R1-Zero RL
- Cold-start data
- R1 SFT and RL
- Reasoning samples
- Dense-model fine-tuning
Models and evaluation
- R1 models
- Distilled checkpoints
- Benchmark evaluation
Other components
- Researcher
- Research community
Connections 12
- Researcher → V3 base model: starts from
README.md - V3 base model → R1-Zero RL: trains with RL
README.md - Cold-start data → R1 SFT and RL: seeds training
README.md - V3 base model → R1 SFT and RL: trains with SFT and RL
README.md - R1 SFT and RL → Reasoning samples: generates samples
README.md - Reasoning samples → Dense-model fine-tuning: fine-tunes
README.md - Dense-model fine-tuning → Distilled checkpoints: produces
README.md - R1 SFT and RL → R1 models: releases
README.md - R1 models → Benchmark evaluation: evaluated on
README.md - Distilled checkpoints → Benchmark evaluation: evaluated on
README.md - Research community → R1 models: accesses
README.md - Research community → Distilled checkpoints: downloads
README.md