# Inkling

Sample, evaluate, and post-train Inkling and Inkling-Small, Thinking Machines Lab's...

- Repository: https://github.com/thinking-machines-lab/tinker-cookbook (4231 stars)
- Category: Writing
- License: Apache-2.0
- Last updated: 2026-10-30
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add thinking-machines-lab/tinker-cookbook --skill inkling -g -a claude-code -y`
Codex: `npx skills add thinking-machines-lab/tinker-cookbook --skill inkling -g -a codex -y`
Cursor: `npx skills add thinking-machines-lab/tinker-cookbook --skill inkling -g -a cursor -y`
Gemini CLI: `npx skills add thinking-machines-lab/tinker-cookbook --skill inkling -g -a gemini-cli -y`
Claude Code plugin marketplace: `/plugin marketplace add thinking-machines-lab/tinker-cookbook`

## Description

Sample, evaluate, and post-train Inkling and Inkling-Small, Thinking Machines Lab's models built for Tinker. Use this skill whenever the user mentions Inkling, `thinkingmachines/Inkling`, tml-renderers, `tml_v0` / `TmlV0Renderer`, or thinking/reasoning effort, and whenever they are choosing a model, building training data, running evals, setting up SFT or RL, handling parse errors, or working with audio or image inputs for an Inkling model. Inkling has requirements that differ from other Tinker models (mandatory effort conditioning, its own renderer and tokenizer, a learning rate you calibrate yourself), so load this skill before writing any Inkling code.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# Inkling

[Inkling](https://thinkingmachines.ai/inkling/) is Thinking Machines Lab's open-weight model family built for Tinker: general-purpose models that code, reason, call tools, and accept image and audio input.

Inkling is a Mixture-of-Experts transformer with 975B total parameters and 41B active. Inkling-Small is an efficient sibling at 276B total and 12B active, reaching comparable performance to Inkling at a quarter of its size. Both offer native reasoning over audio and images, variable thinking effort, a context window of up to 1M tokens, and well-rounded performance across a range of benchmarks.

Inkling-Small's efficiency makes it a reasonable default for most tasks, and a natural fit for workloads where cost and latency matter, such as coding, using LLMs to grade, or generating synthetic data for other models. The two share a renderer, tokenizer, and effort interface, so moving between them is a one-line change to the model name, and benchmarking both on your own task is cheap.

Both models are offered as post-trained versions on Tinker, not base models: they arrive instruction-tuned and effort-conditioned, so treat them as starting points for further post-training, SFT, RL, distillation, rather than for continued pretraining.

Working with Inkling differs from other Tinker models in three ways, each covered below: every render needs an explicit thinking-effort value, rendering and tokenization go through `tml-renderers` rather than a Hugging Face chat template, and the learning rate is yours to calibrate.

## Setup

Inkling renders through the standalone [`tml-renderers`](https://pypi.org/project/tml-renderers/) package, included in the default installation together with the required `torch>=2.10`.

The rest: https://github.com/thinking-machines-lab/tinker-cookbook/blob/HEAD/skills/inkling/SKILL.md