Inkling Small (xHigh): benchmark results
Provider: Thinking Machines. Released 2026-07-30. Access: Open.
Unified ELO 1631 ± 1, rank #443 of 3078 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Epoch AI - GPQA Diamond | 88.51 | Accuracy (%) | 84 |
| OTIS Mock AIME 2024-25 | 90 | Accuracy (%) | 78.4 |
| FrontierMath - Tiers 1-3 (v2) | 46.32 | Accuracy (%, 285 private v2 problems) | 67.2 |
| ARC-AGI-1 | 84 | Accuracy (%) | 64.3 |
| ARC-AGI-2 | 40.14 | Accuracy (%) | 63.7 |
| Chess Puzzles (Epoch AI) | 18 | Accuracy (%) | 56.3 |
| FrontierMath - Tier 4 (v2) | 17.07 | Accuracy (%, 41 private v2 problems) | 34.3 |
| SimpleQA Verified | 19.1 | Accuracy (%) | 15.2 |
| Epoch AI - Mystery Game Puzzles | 6 | Score | 8.3 |
| Chartography | 9.1 | Score (%) | 3.5 |
| HANDBOOK.md Agents | 1.2 | Score (%) | 3.5 |
Interactive version: theaggregate.ai/model?slug=inkling-small-xhigh · How It Works · Data refreshed daily, snapshot 2026-09-19.