Qwen·released 2026-04-201 source
Qwen 3.6 Max (Preview) benchmark scores: 11 benchmarks tracked. 55% of its scores are independently reproduced — the rest are self-reported or unverified.
Leads on 0 of 11 benchmarks·best result 91.1 on OTIS Mock AIME 2024-2025 (reproduced)·6 of 11 independently reproduced·$1.3/$7.8 per M tokens
Source: models.dev · See every model’s pricing →
| Benchmark | Score | Evidence status | Measured |
|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 91.1accuracy (%) | reproduced· optimizedT1 | 2026-04-20Epoch AI |
| GPQA diamond | 83.16accuracy (%) | reproduced· optimizedT1 | 2026-04-20Epoch AI |
| DTBench | 78.67accuracy (%) | unverifiedT1 | 2026-04-20https://conceptualreasoning.ai/dtbench |
| SWE-Bench verified | 76.65accuracy (%) | reproduced· optimizedT1 | 2026-04-20Epoch AI |
| SimpleBench | 55.6accuracy (%) | unverifiedT1 | 2026-04-20SimpleBench Leaderboard |
| SimpleQA Verified | 52accuracy (%) | unverifiedT2 | 2026-04-20 |
| LMCA | 50.04accuracy (%) | unverifiedT1 | 2026-04-20https://conceptualreasoning.ai/lmca |
| FrontierMath-2025-02-28-Private | 40.53accuracy (%) | reproduced· optimizedT1 | 2026-04-20Epoch AI |
| Chess Puzzles | 15.82accuracy (%) | reproducedT1 | 2026-04-20Epoch AI |
| Mystery Game Puzzles | 10.77accuracy (%) | unverifiedT2 | 2026-04-20 |
| FrontierMath-Tier-4-2025-07-01-Private | 6.94accuracy (%) | reproducedT1 | 2026-04-20Epoch AI |
Claims drawn from cited facts, not live model generation.
This model originates from Qwen, placing it within that developer's line of work. Its vintage is 2026, making it a recent entry whose scale falls outside what these facts specify; no parameter count is offered, so its efficiency-versus-headroom trade-off cannot be stated.
3 cited facts
This model is scored on 11 tracked benchmarks, and it currently holds no top score on any of them. Its record has been independently reproduced, meaning outside evaluation backs the reported scores up rather than leaving them as unverified vendor claims.
3 cited facts
On average it trails the leader by 26.8 points under the disclosed harness, a wide gap that reads as well behind the leaders rather than competitive or front-of-pack. It holds the top score on 0 benchmarks, meaning there is no category leadership yet.
2 cited facts
5 facts cross-checked across data sources: 5 single-source.
Source facts, citations, and refresh stamp for this record.
Sources: curated_capability_claims, epoch_benchmarks, modelsdev_models
Qwen 3.6 Max (Preview) is an AI model developed by Qwen, released 2026-04-20. Its benchmark record below tags every score as independently reproduced, self-reported, or unverified.
Qwen 3.6 Max (Preview) has recorded scores on 11 benchmarks, each shown with its evidence status.
Qwen 3.6 Max (Preview) has recorded scores on 11 benchmarks — GPQA diamond, OTIS Mock AIME 2024-2025, DTBench, LMCA, SimpleQA Verified, Chess Puzzles, and 5 more. The full table above shows each score with its evidence status.
6 of 11 recorded scores (55%) are independently reproduced rather than self-reported by the lab.
Qwen 3.6 Max (Preview) has 11 tracked claims: 6 independently reproduced, 5 unverified.
Listed API pricing: $1.3 per million input tokens, $7.8 per million output tokens. See the pricing block for the full breakdown.