ZMIME
Thinking Machines · Released Jul 30, 2026

Inkling Small

Multimodal MoE reasoning model (276B total, 12B active) for text, image, and audio.

  • Open weights
  • Reasoning
  • Tool calling
  • Vision
  • Audio input
Capability
150.8ECI · #65 of 158
Input
$0.58per 1M tokens
Output
$1.44per 1M tokens
Context
1.05M1.05M max output
Pricing

What it costs

Input
$0.58per million tokens
Output
$1.44per million tokens
Cached input
$0.116per million tokens
Blended (3:1)
$0.795pricier than 51% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$1.88
Team assistant25M input + 5M output tokens$21.70
Production app250M input + 50M output tokens$217.00

Official Thinking Machines API price, as listed on models.dev.

Capability

Independent benchmarks

150.8Epoch Capabilities Index
#65 of 158 scored models
68Median 147174

The shaded band is Epoch AI’s confidence range (149.0–152.6); the tick marks the median scored model.

  • GPQA DiamondGraduate-level science questions · xhigh setting88.5%
  • FrontierMath Tiers 1–3Research-level mathematics · xhigh setting46.3%
  • OTIS Mock AIME 2024–2025Competition mathematics · xhigh setting90.0%
  • SimpleQA VerifiedShort factual questions · xhigh setting19.1%

Scores from Epoch AI, run independently of Thinking Machines.

Specifications

The details

Lab
Thinking Machines
Released
Jul 30, 2026
Context window
1,048,576 tokens
Max output
1,048,576 tokens
Inputs
Text, Images, Audio
Output
Text
Reasoning
Adjustable effort low · medium · high · xhigh · max
Tool calling
Yes
Structured output
No
API model ID
thinkingmachines/Inkling-Small:peft:262144on Thinking Machines
Availability
12 API providerslisted on models.dev

Thinking Machines model documentation

Same lab

More from Thinking Machines

Questions

About Inkling Small

How much does Inkling Small cost?

Inkling Small costs $0.58 input / $1.44 output per million tokens (official Thinking Machines API price). Cached input is $0.116 per million tokens. At a 3:1 input-to-output mix that is $0.795 per million tokens, more expensive than 51% of the 469 priced models we track.

What is the context window of Inkling Small?

Inkling Small accepts up to 1,048,576 tokens per request and can write up to 1,048,576 tokens in one response.

How good is Inkling Small?

Epoch AI gives Inkling Small a Capabilities Index score of 150.8 (likely range 149.0–152.6), ranking it #65 of 158 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 88.5%, FrontierMath Tiers 1–3 46.3%, OTIS Mock AIME 2024–2025 90.0%, SimpleQA Verified 19.1%.

Is Inkling Small open source?

Yes. Thinking Machines publishes the weights under the Apache-2.0 licence, so it can be downloaded and self-hosted.

What inputs does Inkling Small support?

Inkling Small accepts text, images and audio and replies in text. It is a reasoning model with low, medium, high, xhigh and max effort settings, supports tool calling.

When was Inkling Small released?

Thinking Machines released Inkling Small on Jul 30, 2026.

What are the best alternatives to Inkling Small?

The closest current models from other labs on capability, price and release date are Qwen3.8 27B (Alibaba (Qwen), ECI 152.7, $0.40 / $2.42), Seed 2.1 Turbo (ByteDance Seed, $0.445 / $2.23), Kimi K2.7 Code (Moonshot AI, ECI 151.2, $0.95 / $4.00) and Gemini 3.5 Flash Lite (Google, ECI 145.7, $0.30 / $2.50).