Overview
Tülu 3 publishes the full post-training pipeline (supervised fine-tuning, preference tuning and reinforcement learning with verifiable rewards) so others can reproduce or adapt it.
Highlights
- 8B, 70B and 405B versions
- Open post-training data and code
- Introduced RL with verifiable rewards (RLVR)
- Based on Llama 3.1
Typical uses
Research on post-training and transparent assistants.
Things to consider
- Llama license terms apply.
- Download sizes refer to the default quantized versions in the Ollama library; full-precision weights are larger.
Specifications are taken from the model card on Hugging Face, the Ollama library and the provider’s announcement. Benchmark results depend on evaluation settings; see how AI benchmarks work.