released 2026-07-151 source
Inkling-Small benchmark scores: 10 benchmarks tracked. 60% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 10 benchmarks·best result 89.99 on OTIS Mock AIME 2024-2025 (reproduced)·6 of 10 independently reproduced
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 89.99accuracy (%) | reproduced· optimizedT1 | 2026-07-15Epoch AI |
| GPQA diamond | 84.68accuracy (%) | reproduced· optimizedT1 | 2026-07-15Epoch AI |
| ARC-AGI | 84accuracy (%) | unverified· optimizedT1 | 2026-07-15https://arcprize.org/leaderboard |
| FrontierMath-Tiers-1-3-v2-Private | 46.32accuracy (%) | reproducedT1 | 2026-07-15Epoch AI |
| ARC-AGI-2 | 40.14accuracy (%) | unverifiedT2 | 2026-07-15 |
| SimpleQA Verified | 19.1accuracy (%) | unverifiedT2 | 2026-07-15 |
| FrontierMath-Tier-4-v2-Private | 17.07accuracy (%) | reproducedT1 | 2026-07-15Epoch AI |
| Chess Puzzles | 13.72accuracy (%) | reproducedT1 | 2026-07-15Epoch AI |
| ProofBench | 6accuracy (%) | unverifiedT1 | 2026-07-15https://www.vals.ai/benchmarks/proof_bench |
| Mystery Game Puzzles | 0accuracy (%) | reproducedT1 | 2026-07-15Epoch AI |
Claims drawn from cited facts, not live model generation.
This model dates to 2026, making it a current-generation release rather than a legacy one.
1 cited fact
This model is scored on 10 tracked benchmarks, and it holds no current top score on any of them. Its record can be trusted: the scores have been independently reproduced, meaning outside evaluation backs them up rather than resting on vendor claims alone.
3 cited facts
Its mean gap to the best recorded score across its tracked benchmarks is 51.24 points.
1 cited fact
1 facts cross-checked across data sources: 1 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: epoch_benchmarks
Inkling-Small is an AI model released 2026-07-15. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Inkling-Small has recorded scores on 10 benchmarks, each shown with its evidence status.
Inkling-Small has recorded scores on 10 benchmarks — GPQA diamond, OTIS Mock AIME 2024-2025, SimpleQA Verified, Chess Puzzles, ARC-AGI, ARC-AGI-2, and 4 more. The full table above shows each score with its evidence status.
6 of 10 recorded scores (60%) are independently reproduced rather than self-reported by the lab.
Inkling-Small has 10 tracked claims: 6 independently reproduced, 4 unverified.