Can Inkling Small run locally?

Yes — the weights are open and the model card lists recipes for SGLang, vLLM, Unsloth, TokenSpeed, and Hugging Face, so you can self-host it. Bear in mind the total parameter count is 276B, so you will need substantial hardware even though only 12B parameters are active per token.

What does Inkling Small cost and what is its context window?

Via API it is priced at $0.45 per million input tokens and $1.20 per million output tokens, with a 524,288-token context window.

How does Inkling Small differ from the full Inkling model?

Inkling Small is the smaller, more efficient member of the family — lower active parameter count and lower API price — while the full Inkling model sits above it in the lineup. Thinking Machines has not published a detailed capability comparison between the two.

Inkling Small

5840°

#llm #thinkingmachines #vision #audio #reasoning #agents

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

Thinking Machines | 524,288 token context | $0.45/M input, $1.20/M output. Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters out of 276B total. It is positioned as the smaller, more efficient member of the Inkling family and is suited for reasoning, coding, agentic workflows, retrieval-augmented generation, instruction following, and multilingual conversation.

Guides for this model

Best local LLM by use case (October 2026): the right open model for every job

How much VRAM do I actually need for local AI?

Open weights vs open source vs API-only: what you are actually allowed to do

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Inkling Small is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 12B active parameters drawn from a 276B total pool. It accepts text, image, and audio inputs, runs locally, and targets developers building agentic systems, coding assistants, and RAG pipelines. The price is low enough that API use stays cheap even at moderate scale.

Known for

Natively handles text, image, and audio in a single model
MoE architecture: 12B active params keep inference costs down
512K token context window — unusually large for its class
Designed for agentic and tool-use workflows out of the box
Open weights support local deployment via SGLang, vLLM, and Unsloth

Watch out for

Audio input requires WAV at 16kHz; clips over two minutes may degrade in quality.
276B total parameters means local deployment needs serious hardware despite the low active count.
Licence is Acceptable Use, not fully permissive — check Thinking Machines' terms before commercial use.

Best for

Prompts to try

Debug a broken agent loop

I have a Python agent that calls a search tool, parses the result, and decides whether to call again or return a final answer. It sometimes loops indefinitely. Here is the code: [paste yours]. Identify the likely cause of the infinite loop, explain why it happens, and rewrite the relevant section with a fix and a max-iterations guard.

Reason through a planning problem

A small team of four must complete six tasks before a Friday deadline. Tasks A and D must finish before B can start. Task C depends on B. Tasks E and F are independent. Two people can work in parallel at most. Map out the critical path, identify the earliest possible finish day assuming each task takes one day, and flag any scheduling risks.

Multilingual instruction extraction

The following customer support message is written in a mix of Tagalog and English. Extract the core complaint, the product mentioned, and the action the customer is requesting. Return your answer as a JSON object with keys: complaint, product, requested_action. Message: 'Hindi gumagana yung headset na binili ko last week, parang may sira yung left ear. Can you replace it or give a refund?'

Compare with

Inkling
Open-weights model
524,288 token context · $1.00/M input, $4.05/M output
GPT-6 Luna
API-only model
1,050,000 token context · $0.10/M input, $0.50/M output
GPT-5.6 Luna
API-only model
1,050,000 token context · $0.20/M input, $1.20/M output
Claude Opus 5.5
API-only model
1,000,000 token context · $4.00/M input, $20.00/M output

More text models from thinkingmachines

Questions people ask

Can Inkling Small run locally?: Yes — the weights are open and the model card lists recipes for SGLang, vLLM, Unsloth, TokenSpeed, and Hugging Face, so you can self-host it. Bear in mind the total parameter count is 276B, so you will need substantial hardware even though only 12B parameters are active per token.
What does Inkling Small cost and what is its context window?: Via API it is priced at $0.45 per million input tokens and $1.20 per million output tokens, with a 524,288-token context window.
How does Inkling Small differ from the full Inkling model?: Inkling Small is the smaller, more efficient member of the family — lower active parameter count and lower API price — while the full Inkling model sits above it in the lineup. Thinking Machines has not published a detailed capability comparison between the two.

Captain's verdict
A rare small model that genuinely handles text, image, and audio together — worth a look if you are building multimodal agents and want open weights.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Not this creator?

Get your own badge

How this score adds up

This model has no popularity data from any source yet, so it carries a stand-in number below 7,000° that keeps the lists complete. It never ranks above a model with data. The breakdown appears here as soon as a source reports it.

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

Similar Models

AI Models for Image, Video & Text Generation

Inkling Small is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm thinkingmachines vision audio reasoning agents coding multilingual

Browse by Model Family