Thinking Machines model benchmarks.

Compare Thinking Machines models across answer quality, coding, complex tasks, speed, price, and context size.

Thinking Machines

Quality

Highest overall scores

  1. 1Thinking Machines: Inkling40.7
  2. 2Thinking Machines: Inkling Small40.2

Response speed

Fastest output in this sample

  1. 1Thinking Machines: Inkling136 t/s
  2. 2Thinking Machines: Inkling Small0 t/s

Input price

Lowest price per 1M tokens

  1. 1Thinking Machines: Inkling Small$0.50/1M
  2. 2Thinking Machines: Inkling$1.00/1M

Highlights

Quality

Overall capability · Higher is better

Speed

Output tokens per second · Higher is better

Input price

USD per 1M tokens · Lower is better

Quality benchmarks

Compare Thinking Machines model capability.

Performance and price

Compare practical tradeoffs.

All Thinking Machines models.

2 models in the current catalogue.

ModelLinks
Thinking Machines: Inkling40.7136 t/s1M$1.00$4.05
Boring technical details Creator of the model
Thinking Machines: Inkling Small40.2—524.3K$0.50$1.20
Boring technical details Creator of the model

Thinking Machines technical benchmarks.

Compare named evaluation records available for Thinking Machines models. Missing results remain blank rather than being estimated.

BenchmarkAreaInklingInkling SmallSource
Intelligence IndexoverallQuality40.740.2OpenRouter / artificial-analysis
Coding Indexcoding52.152.9OpenRouter / artificial-analysis
Agentic Indexagentic32.330.8OpenRouter / artificial-analysis
GPQAGPQA Diamond87.289.5OpenRouter model benchmarks
Humanity's Last ExamHLE29.731.6OpenRouter model benchmarks
AA-LCRAA-LCR63.363OpenRouter model benchmarks
GDPval-AAGDPval-AA36.838.4OpenRouter model benchmarks
CritPtCritPt5.48.3OpenRouter model benchmarks
SciCodeSciCode46.148.7OpenRouter model benchmarks
AA-Omniscience AccuracyAA-Omniscience Accuracy4030.5OpenRouter model benchmarks
AA-Omniscience Non-Hallucination RateAA-Omniscience Non-Hallucination Rate36.943.1OpenRouter model benchmarks