Skip to content
ShilpBot
AI Models

Thinking Machines Inkling 2 Small (Free): Specs, Limits and How to Use It

Thinking Machines Inkling 2 Small free: open-weight multimodal mixture-of-experts reasoning model, text/image input, 1M context. Specs, limits, test...

By ShilpBot Team · Published · 3 min read

Inkling 2 Small is an open-weight multimodal mixture-of-experts reasoning model from Thinking Machines Lab. It has 17 billion active parameters out of 594 billion in total.

The free version is useful for testing before you pay. A paid listing of the same model also exists, so you can prototype on the free tier and then move to the paid route when you need predictable availability.

Specifications checked against the public OpenRouter catalogue on 2026-11-16. See the live model page for current limits.

Thinking Machines Inkling 2 Small quick facts

DetailValue
DeveloperThinking Machines Lab
Model IDthinkingmachines/inkling-2-small:free
Added to the catalogue2026-11-09
TypeOpen-weight multimodal mixture-of-experts reasoning model
Parameters17B active / 594B total
Context windowabout 1M tokens (1,048,576)
Maximum output131,072 tokens
Input typestext, image
PriceFree (paid listing: $0.30 in / $1.00 out per 1M tokens)

Feature support from the catalogue: Tool calling: listed; JSON / structured output: not listed; Adjustable reasoning effort: listed.

What Thinking Machines Inkling 2 Small is good for

  • Image-plus-text reasoning, such as reading a chart and then explaining the trend.
  • Prototyping with an open-weight model before deciding whether to self-host later.
  • Experiments with reasoning effort, since the model exposes that parameter.

Limits of Thinking Machines Inkling 2 Small

  • The catalogue does not list structured-output (JSON schema) support for this model, so validate any JSON you request.
  • Open weights do not mean the free hosted endpoint is private. Treat prompts as potentially logged.
  • Free availability can be rate limited at busy times.

How Thinking Machines Inkling 2 Small compares

  • NVIDIA Nemotron 3.5 Super VL: It has a larger context window (1,048,576 against 1,000,000 tokens); both cap output at 131,072 tokens; it does not accept video input; it activates 17B parameters per token against 12B.
  • Dots Studio Dots3-Note Preview: It has a larger context window (1,048,576 against 512,000 tokens); its output cap is 131,072 tokens against 460,800; it activates 17B parameters per token against 16B.
  • Baidu ERNIE 5.5 Flash Preview: It has a larger context window (1,048,576 against 1,024,000 tokens); its output cap is 131,072 tokens against 921,600; it also accepts image input.

Calling Thinking Machines Inkling 2 Small from code

Use the model ID thinkingmachines/inkling-2-small:free with any OpenAI-compatible client pointed at OpenRouter:

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "thinkingmachines/inkling-2-small:free", "messages": [{"role": "user", "content": "Explain what you can do in two sentences."}]}'

Prompts that show what Thinking Machines Inkling 2 Small can do

  1. Upload a bar chart and ask for three conclusions plus one thing the chart cannot tell you.
  2. Ask it to plan a database migration in steps, then to list the risks of each step.
  3. Give it a tricky logic puzzle with reasoning effort set to low, then high, and compare.

What Thinking Machines Inkling 2 Small costs

The free listing costs nothing, but the same model also has a paid listing at about $0.30 per million input tokens and $1.00 per million output tokens. As a rough guide, sending 1 million input tokens and receiving 100,000 output tokens would cost about $0.40 on the paid listing. Use the free tier to test, and move to the paid route if you need steadier availability. Free pricing is set by the provider and can change at any time, so check the model page before you plan around it.

Keep reading

All free models are compared in our free AI chat models roundup. New to terms like mixture-of-experts? Start with how neural networks learn.

Frequently asked questions

Is there a paid version of Inkling 2 Small?

Yes. The catalogue lists a paid version at about $0.30 per million input tokens and $1.00 per million output tokens.

Is Thinking Machines Inkling 2 Small free, and what are its limits?

As of 2026-11-16 the catalogue lists it at zero price. About 1M tokens (1,048,576) context window, up to 131,072 output tokens, model ID thinkingmachines/inkling-2-small:free. Free use can be rate limited and providers may change it.

#Thinking Machines Inkling 2 Small#Thinking Machines Lab free model#free AI model#AI chat model#reasoning model#Kilo Gateway#OpenRouter