Guide for installing Tinker, setting up the environment, getting an API key, and verifying everything works. Use when the user is getting started, setting up their environment, or troubleshooting installation issues.
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
Get Tinker and tinker-cookbook running from scratch.
docs/install.mdx — Official installation guideCONTRIBUTING.md — Development setupREADME.md — Project overviewexport TINKER_API_KEY=<your-key>
Add to your shell profile (.bashrc, .zshrc) for persistence.
pip install tinker
This gives you:
TrainingClient, SamplingClient, low-level training/sampling APIstinker or python -m tinker for management tasksgit clone https://github.com/thinking-machines-lab/tinker-cookbook.git
cd tinker-cookbook
pip install -e .
Or with dev dependencies (for contributing):
uv sync --extra dev
pre-commit install
import tinker
service_client = tinker.ServiceClient()
training_client = service_client.create_lora_training_client(
base_model="meta-llama/Llama-3.2-1B", rank=32,
)
info = training_client.get_info()
print(info) # Should print model info
# Supervised learning
python -m tinker_cookbook.recipes.sl_basic
# Reinforcement learning
python -m tinker_cookbook.recipes.rl_basic
| Variable | Purpose |
|---|---|
TINKER_API_KEY | Required — authenticates with Tinker service |
HF_TOKEN | Optional — access gated HuggingFace models (Llama, etc.) |
HF_TRUST_REMOTE_CODE | Optional — allow custom tokenizer code |
WANDB_API_KEY | Optional — log to Weights & Biases |
TINKER_API_KEY not set: Export the key in your shell or .env fileHF_TOKEN for gated models (e.g., Llama)pip install -e . was run from the repo rootuv not found: Install with curl -LsSf https://astral.sh/uv/install.sh | shNo reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Guide for checkpointing — saving, loading, and resuming training with CheckpointRecord. Use when the user asks about saving weights, resuming training, checkpoint management, or the checkpoint lifecycle.
The English summary is being prepared. Showing the original description.
Guide for testing conventions and CI pipelines — unit tests, integration smoke tests, pytest markers, and GitHub Actions workflows. Use when the user asks about testing, CI, running tests, or adding tests for a recipe.
The English summary is being prepared. Showing the original description.
Guide for using completers — TokenCompleter and MessageCompleter for text generation during RL rollouts and evaluation. Use when the user asks about generating text, completing messages, or using completers in RL environments.
The English summary is being prepared. Showing the original description.
Guide for contributing to the tinker-cookbook repo — development setup, code style, type checking, PR process, and design conventions. Use when the user asks about how to contribute, set up the dev environment, code style, or project conventions.
The English summary is being prepared. Showing the original description.
Guide for dataset construction — SupervisedDatasetBuilder, RLDatasetBuilder, ChatDatasetBuilder, and custom dataset creation from JSONL, HuggingFace, or conversation data. Use when the user asks about datasets, data loading, data preparation, or custom data formats.
The English summary is being prepared. Showing the original description.
Set up and run knowledge distillation (on-policy, off-policy, or multi-teacher) from a teacher model to a student model using the Tinker API. Use when the user wants to distill knowledge, compress models, or train a student from a teacher.
The English summary is being prepared. Showing the original description.