ModelPriceWatch.com
Last scan 2026-08-01 Models tracked 187 Providers 28 Cheapest paid Granite 4.0 Micro $0.017/Mtok in Every price links to its source

Inkling Small

by Thinking Machines · 276B total / 12B active parameters

Current new · 2d Intro price budget open weights cheap tier

Today's price · per 1M tokens

Input

$0.300

per 1M tokens

Output

$1.20

per 1M tokens

Blended

$0.525

blended $/1M — 3:1 weighted input:output

Cached input

$0.060

20% of input — prompt caching

Source: official Thinking Machines pricing · read Aug 1, 2026 MODELPRICEWATCH.COM · 2026-08-01

Overview

The lighter Inkling: a 276B-total / 12B-active MoE that matches its larger sibling on many benchmarks at a quarter of the size, with the same native text/image/audio reasoning and variable thinking effort. Open weights on Hugging Face, sold through Tinker serverless inference at $0.30/$1.20 per 1M tokens.

Run it yourself

Deploy this open model on rented GPUs

Open weights mean you can self-host instead of paying per-token API prices. These platforms let you serve it on demand — often cheaper at scale.

Partner links — we may earn a commission if you sign up, at no cost to you. We only list platforms we'd recommend regardless.

Capabilities

struck through = not supported
Input 3/5
Text ✓ Image ✓ Audio ✓ Video PDF
Output 1/5
Text ✓ Image Audio Video Embedding
Features 5/9
Prompt caching ✓ Reasoning ✓ Coding Fast inference Long context ✓ Open weights ✓ Multimodal ✓ Web search Realtime

Benchmark performance

accuracy % · higher is better
Percentile vs all tracked models 49.1th4 independent measurements
Percentile per $/Mtok 94.4

We hold no per-benchmark accuracy scores for this model yet, so it has no accuracy average. That is a gap in our coverage, not a sign the model is untested — independent evaluators often publish a composite index for a new model long before releasing its per-benchmark numbers. The percentile above is its standing across the independent composites below.

Independent composite scores
  • AA Intelligence Index: 40.2 (Artificial Analysis composite across reasoning, knowledge and coding evals)
  • AA Agentic Index: 30.8 (Tool use, planning, autonomy)
  • AA-Omniscience Index: -9 (−100–100)
  • LMSYS Chatbot Arena: 1430.5 ELO (Human preference)
How it stacks up
  • Ranks #53 of 136 comparably-measured models by percentile score, across 4 independent measurements
  • Ranks #21 of 136 comparably-tested models by normalized performance per dollar

Source: Artificial Analysis · updated Aug 1, 2026 · See full rankings →

Specifications

Context window
262K tokens
Modality
text, image, audio
Parameters
276B total / 12B active
Open source
Yes — open weights available
Released
Jul 30, 2026
Status
Current
Last updated
Aug 1, 2026
Tags
open-weightsbudgetreasoningmultimodalmoecaching

Availability verified: Aug 1, 2026 — listed on Thinking Machines's own page