thinking-machines-lab

    thinking-machines-lab/tinker-cookbook

    Post-training with Tinker

    machine-learning
    Python
    Apache-2.0
    4.2K stars
    555 forks
    4.2K GitHub watchers
    Updated Oct 26, 2026
    View on GitHub

    Build with Backblaze B2

    SDKs, agent skills, IDE extensions, and reference pipelines from Backblaze Labs. All open source.

    Explore Backblaze Labs

    Loading star history...

    Where it stands

    Rank, growth pace and release cadence relative to everything GStars tracks.

    5 releases in the last 90 days, the latest 5 days ago, with 74 contributors on GitHub.

    Peaked at #26 in the weekly ranking the week of Nov 16, its best in the past year.

    Use Cases & Benefits

    • Provides a Python library with examples and abstractions to fine-tune large language models using the Tinker API.
    • Simplifies complex distributed training workflows by offering reusable recipes and utilities for supervised and reinforcement learning.
    • Use for customizing and fine-tuning conversational AI models on chat datasets with supervised learning techniques.
    • Use for improving language model reasoning and math problem-solving skills through reinforcement learning pipelines.
    • Use for developing reward models and applying reinforcement learning from human feedback to optimize model behavior.

    Should you depend on it?

    License terms, project hygiene and risks, from GitHub metadata and GStars' snapshots.

    License

    Apache 2.0

    Permissive: commercial use, modification and redistribution are allowed with attribution; includes an explicit patent grant from contributors.

    Community files

    • README: present
    • Contributing guide: present

    Watch-outs

    No watch-outs from GStars' checks.

    General guidance from repository metadata, not legal advice. How we score

    From the README

    We provide two libraries for the broader community to customize their language models: tinker and tinker-cookbook.

    Refer to the docs to start from basics. Here we introduce a few Tinker primitives - the basic components to fine-tune LLMs:

    See tinker_cookbook/recipes/sl_loop.py and tinker_cookbook/recipes/rl_loop.py for minimal examples of using these primitives to fine-tune LLMs.

    Excerpt from the thinking-machines-lab/tinker-cookbook README on GitHub (Apache-2.0 license). © its authors.Read the full README on GitHub

    Daily snapshots

    Stars gained per day, and how forks and open issues moved, from GStars' daily snapshots

    Aug 5: +1Aug 6: +3Aug 7: +4Aug 8: +1Aug 9: +4Aug 10: +4Aug 11: +4

    +274 stars over 90 days across 14 snapshot days; best day +4 on Aug 7.

    Forks

    555+61

    Open issues

    139+63

    Compared with peers in Machine Learning

    Fastest-growing repositories GStars tracks in this cohort for the week of Nov 1, by 7-day star growth

    thinking-machines-lab/tinker-cookbook compared with Machine Learning repositories by stars, 7-day star growth, health score and last release
    RepositoryStars+7dHealthLast release
    photoprism/photoprism40.4K—651 day ago
    huggingface/diffusers34.8K—653 days ago
    onnx/onnx21.6K—6331 days ago
    lukas-blecher/LaTeX-OCR16.6K—18—
    milesial/Pytorch-UNet11.7K—18—
    thinking-machines-lab/tinker-cookbook4.2K—605 days ago
    Category median21.6K—63—

    Discover Repositories

    Search across tracked repositories by name or description