Inkling 2 Small

Inkling 2 Small is an open-weight multimodal mixture-of-experts reasoning model from Thinking Machines Lab, with 17B active parameters out of 594B total. It is the first model in the second...

Model card on Hugging Face

Input / output price: $0.30 / $1.00 per 1M
Context: 1M tokens
Released: Nov 5, 2026

Try Respan for free

Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "inkling-2-small".

from openai import OpenAI

client = OpenAI(

    base_url="https://api.respan.ai/api/",

    api_key="YOUR_RESPAN_API_KEY",

)

response = client.chat.completions.create(

    model="inkling-2-small",

    messages=[{"role": "user", "content": "Hello!"}],

)

print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({

  baseURL: "https://api.respan.ai/api/",

  apiKey: "YOUR_RESPAN_API_KEY",

});

const response = await client.chat.completions.create({

  model: "inkling-2-small",

  messages: [{ role: "user", content: "Hello!" }],

});

console.log(response.choices[0].message.content);

curl https://api.respan.ai/api/chat/completions \

  -H "Authorization: Bearer YOUR_RESPAN_API_KEY" \

  -H "Content-Type: application/json" \

  -d '{

    "model": "inkling-2-small",

    "messages": [{"role": "user", "content": "Hello!"}]

}'

Read docs

Providers

4 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.

Provider	Context	Input	Output	Cache read
DeepInfra`deepinfra/thinkingmachines/Inkling-2-Small`	262K	$0.30/M	$1.00/M	$0.03/M
Nebius`nebius/inkling-2-small`	1M	$0.30/M	$1.00/M	$0.06/M
TogetherAI`togetherai/thinkingmachines/Inkling-2-Small`	1M	$0.30/M	$1.00/M	$0.03/M
OpenRouter`openrouter/thinkingmachines/inkling-2-small`	1M	$0.30/M	$1.00/M	$0.03/M

Is a provider down? OpenRouter status

Pricing

Averaged across 4 providers, in USD per 1M tokens after gateway discounts.

Average input: $0.30 per 1M
Average output: $1.00 per 1M
Average cache read: $0.04 per 1M

Cost by volume

Total cost from 0 to 1T tokens.

$0.30/MDeepInfra, Nebius, TogetherAI, and OpenRouter

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

338tokens/sacross providers

Faster than 88% of the other 560 models on the gateway

Time to first token

Lower is better

1.4sacross providers

Faster than 52% of the other 136 models on the gateway

Response latency

Lower is better

11sacross providers

Faster than 16% of the other 650 models on the gateway

Benchmarks

The Respan Index plus 14 published scores from 4 sources, each linked to where it was published.

52.8/ 100

Respan Index#28 of 170

170 ranked models. The shaded band is this model's 95% interval, 40.3 to 64.5. Kimi K3 and Qwen3.8-Max fall inside it, so the gap is within the margin of error.

28Inkling 2 Small52.8
28Kimi K352.8
28Qwen3.8-Max52.8

Published scores

Benchmark	Score	Source	Published
SWE Bench Pro v2-Hardthinkingmachines.ai	86.3%	thinkingmachines.ai	Nov 5, 2026
DeepSWE v1.1thinkingmachines.ai	72.1%	thinkingmachines.ai	Nov 5, 2026
Terminal-Bench 4.0Vals AI	12.6%	Vals AI	Nov 13, 2026

Benchmark	Score	Source	Published
BrowseComp (context compaction)thinkingmachines.ai	90.2%	thinkingmachines.ai	Nov 5, 2026
MCP Atlasthinkingmachines.ai	85.4%	thinkingmachines.ai	Nov 5, 2026
APEX-AgentsMercor	46.8%	Mercor	Nov 13, 2026

Benchmark	Score	Source	Published
AA-LCR v1.1thinkingmachines.ai	83.3%	thinkingmachines.ai	Nov 5, 2026
ARC-AGI-2thinkingmachines.ai	81.7%	thinkingmachines.ai	Nov 5, 2026
Artificial Analysis Intelligence Indexthinkingmachines.ai	30.6 · #4 of 4	thinkingmachines.ai	Nov 5, 2026

Benchmark	Score	Source	Published
MMMU-Pro (standard/vision)thinkingmachines.ai	81.6%	thinkingmachines.ai	Nov 5, 2026

Benchmark	Score	Source	Published
LMArena WebDevLMArena	1647 Elo · #20 of 120	LMArena	Nov 11, 2026
LMArena EloLMArena	1412 Elo · #109 of 181	LMArena	Nov 11, 2026
LMArena AgentLMArena	0.0251 · #20 of 63	LMArena	Nov 11, 2026

Benchmark	Score	Source	Published
SAGE 1.1Vals AI	88.6%	Vals AI	Nov 13, 2026

How the Respan Index works

Apps

Use Inkling 2 Small in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

Inkling 2 Small FAQ

Short answers, built from the data on this page.

Inkling 2 Small is a model from Thinking Machines, released on Nov 5, 2026. Inkling 2 Small is an open-weight multimodal mixture-of-experts reasoning model from Thinking Machines Lab, with 17B active parameters out of 594B total. It is the first model in the second...

On the Respan gateway, Inkling 2 Small costs $0.30 per 1M input tokens and $1.00 per 1M output tokens on every provider, after gateway discounts.

Inkling 2 Small is available through 4 providers: DeepInfra, Nebius, TogetherAI, and OpenRouter.

Inkling 2 Small accepts up to 1M tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://api.respan.ai/api/ with your Respan API key and set the model to "inkling-2-small". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Inkling 2 Small scores 52.8 out of 100 on the Respan Index v1.1, ranking #28 of 170 models. By category: Coding 29.8, Agents 63.1, Reasoning 63.0, Knowledge & language 52.6 (estimated), and Human preference 57.3. The Respan Index combines independent and lab-reported benchmark results into one score, using benchmark data collected by Respan.

Inkling 2 Small has 14 published benchmark scores, including SWE Bench Pro v2-Hard 86.3%, BrowseComp (context compaction) 90.2%, and AA-LCR v1.1 83.3%. Each score links to where it was published; the data is collected by Respan.