Tinker Cookbook is a collection of libraries and examples for fine-tuning language models through the Tinker training service, which handles distributed training behind an API. It is used by researchers and developers for language-model post-training and experimentation, with examples built around common fine-tuning tasks. The catalogue entries provide agent workflows and integrations for using the cookbook.
Borrowing it
Nothing to install: this file belongs to thinking-machines-lab/tinker-cookbook. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/thinking-machines-lab/tinker-cookbook/main/AGENTS.mdgit clone --depth 1 https://github.com/thinking-machines-lab/tinker-cookbookWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/thinking-machines-lab/tinker-cookbook/agents-md)<a href="https://agentmods.dev/instructions/thinking-machines-lab/tinker-cookbook/agents-md"><img src="https://agentmods.dev/badge/instructions/thinking-machines-lab/tinker-cookbook/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.01835 | $0.01835 |
| Opus 5 | $0.00918 | $0.00918 |
| Sonnet 5 | $0.00367 | $0.00367 |
| Haiku 4.5 | $0.00184 | $0.00184 |
Grade A, and why
tinker-cookbook AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tinker Cookbook Agent Guide
Quick reference for agents working on tinker-cookbook. Detailed guidance is in the skills under skills/.
tinker-cookbook is a client library with training and eval code built on the Tinker service (hosted by Thinking Machines Lab) and the Tinker SDK (a separate repo with just the API). You author training/eval loops that run on a CPU machine; Tinker executes the heavy GPU work.
Skills: This repo ships three Claude Code skills in skills/: research (SFT, RL, DPO, distillation, evaluation, model selection, experiment methodology), debug (performance, correctness, renderer, and error triage), and inkling (thinking effort, rendering, post-training, and multimodal input for Inkling series models). Install via /plugin marketplace add thinking-machines-lab/tinker-cookbook, then use /tinker:research, /tinker:debug, or /tinker:inkling.
Composing Types
Agents often struggle with the nested type hierarchy.
Core types:
Datum=model_input(ModelInput) +loss_fn_inputs(dict of TensorData)ModelInput= list of chunks (EncodedTextChunk, ImageChunk)TensorData= wrapper for numpy/torch arrays with shape info
Helper functions (use these instead of manual construction):
datum_from_model_input_weights(model_input, weights, max_length)- SL datum creation (supervised/common.py)conversation_to_datum(messages, renderer, max_length, train_on_what)- Full pipeline (supervised/data.py)renderer.build_supervised_example(messages)- Returns (ModelInput, weights)ModelInput.from_ints(tokens)- Create from token listTensorData.from_numpy(arr)/TensorData.from_torch(tensor)- Wrap arrays
Architecture
Builder pattern: Config objects are chz dataclasses (SupervisedDatasetBuilder, RLDatasetBuilder, EnvGroupBuilder). They expose .build()/__call__() returning runtime objects.
Key code locations:
- SL:
tinker_cookbook/supervised/train.py - RL:
tinker_cookbook/rl/train.py - DPO:
tinker_cookbook/preference/train_dpo.py - Renderers:
tinker_cookbook/renderers/ - Completers:
tinker_cookbook/completers.py - RL types:
tinker_cookbook/rl/types.py - Rollout strategies:
tinker_cookbook/rl/rollout_strategy.py(FailFast, RetryOnFailure) - Logging:
tinker_cookbook/utils/logtree.py,tinker_cookbook/rl/rollouts.py - Recipes:
tinker_cookbook/recipes/
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 111 lines · 1,835 tokens per session scan A 1ccc02bb74c1
tinker-cookbook AGENTS.md is an instructions file published in the GitHub repository thinking-machines-lab/tinker-cookbook (4,097 stars, last pushed yesterday), licensed Apache-2.0. It adds 1,835 tokens to every session, about $0.0092 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.