thinkingmachinesListed Jul 27, 2026

Inkling-Small

Inkling-Small is a Mixture-of-Experts transformer with 276B total parameters, 12B active, trained on NVIDIA GB300 NVL72 systems. Like Inkling, it features native reasoning over audio and images, variable thinking effort

Context window524.3K
Max output262.1K
Provider endpoints4
Benchmark results0
Measured performance

Benchmark results

All leaderboards ↗
—

No published results in this index yet

Scores appear here when an evaluation matches this exact model.

Access & cost

Prices across providers

4 quotes

Your token mix

USD · listed token rates

DeepInfra · via OpenRouterdeepinfra/fp8 · quantization=fp8
$0.75

OpenRouter

Catalog quote
Input
$0.45
Output
$1.20
Cached input
$0.10
USD / 1M tokens
524.3K context262.1K max outputPrice source ↗Checked Oct 17, 2026
Endpoint IDthinkingmachines/inkling-small

deepinfra

via Hugging Face
Input
$0.45
Output
$1.20
USD / 1M tokens
524.3K contextPrice source ↗Checked Oct 17, 2026
Endpoint IDthinkingmachines/Inkling-Small
Input
$0.45
Output
$1.20
USD / 1M tokens
524.3K contextPrice source ↗Checked Oct 17, 2026
Endpoint IDthinkingmachines/Inkling-Small

DeepInfra

via OpenRouter

deepinfra/fp8 · quantization=fp8

Input
$0.45
Output
$1.20
Cached input
$0.10
USD / 1M tokens
524.3K context262.1K max outputPrice source ↗Checked Oct 17, 2026
Endpoint IDthinkingmachines/inkling-small::deepinfra/fp8

Model specifications

Developer
thinkingmachines
Modality
Language
Context
524,288 tokens
Maximum output
262,144 tokens
Weights
Open weights
Provider listed
Jul 27, 2026
First discovered
Oct 16, 2026

Exact identities

  • openrouterthinkingmachines/inkling-small
  • huggingfacethinkingmachines/Inkling-Small
  • deepinfrathinkingmachines/Inkling-Small

Sources

Updated Oct 17, 2026

More from thinkingmachines

View all ↗
thinkingmachinesListed Jul 30

Thinking Machines: Inkling Small (free)

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...

1M contextImage inputAudio inputNo reported score
No reported score
$0.00in$0.00out/ 1M tokens · USDThinking Machines · via OpenRouter · 2 endpointsthinkingmachines/nvfp4 · quantization=nvfp4
thinkingmachinesListed Jul 17

Thinking Machines: Inkling (free)

Inkling is an open-weight multimodal mixture-of-experts model from Thinking Machines Lab, with 41B active parameters out of 975B total. It is designed for general-purpose reasoning, coding, agentic and…

1M contextImage inputAudio inputNo reported score
No reported score
$0.00in$0.00out/ 1M tokens · USDThinking Machines · via OpenRouter · 2 endpointsthinkingmachines/nvfp4 · quantization=nvfp4