MMLU-Pro
Broad academic knowledge and reasoning.
Prompting and reasoning budgets can differ between submissions.MODEL BENCHMARKS / CAPABILITY EXPLORERSet your bar.
Set your bar.
Find your model.
Choose the capabilities that matter to you. Set minimum scores and explore which models meet every requirement.
These are published capability scores. Evaluation settings, tool use and reasoning budgets may differ. Use them to build a shortlist; they do not predict quality at your local quantization.
02 / EXPLORE THE LANDSCAPE
View as table Capability, at a glance.
Publisher counts show plotted / available models for your search. Plotting needs both axis scores and follows “Show matches only”.
19meet all minimums4 below a minimum46 missing a required score
Meets all minimumsBelow a minimumMissing required score
03 / INSPECT A MODEL
Meets all minimumsQwen/Qwen3.5-397B-A17B
Qwen · Open weights · 5 of 6 scores reported
What computer can run this model?
Compare memory fit, estimated speed and cost for this exact model.
FP8 · 8K context · 30 tok/s target. Uses FP8 because the current format is not listed for this model. Your other simulation settings are kept.
MMLU-Pro87.8%Accuracy · min 60%
GPQA88.4%Accuracy · min 50%
GSM8K—Not reported
AIME 202693.33%Score
HLE28.7%Accuracy
SWE Verified76.4%Resolved
THE SCORECARD
Every score. The same requirements.
69 models in this selection. Select a model to inspect it above. Scores are percentages; a dash means not reported.
| Model | Your requirements | MMLU-Pro≥ 60% | GPQA≥ 50% | GSM8KNo minimum | AIME 2026No minimum | HLENo minimum | SWE VerifiedNo minimum |
|---|---|---|---|---|---|---|---|
| Missing required score | 88% | — | — | — | 22.2% | 74% | |
| Meets all minimums | 87.8% | 88.4% | — | 93.33% | 28.7% | 76.4% | |
| Meets all minimums | 87.1% | 87.6% | — | 95.83% | 50.2% | 70.8% | |
| Meets all minimums | 86.7% | 86.6% | — | — | 25.3% | 72% | |
| Meets all minimums | 86.1% | 85.5% | — | 90.83% | 24.3% | 72.4% | |
| Meets all minimums | 85.3% | 84.2% | — | 93.33% | 22.4% | 69.2% | |
| Missing required score | 85% | — | — | — | — | — | |
| Meets all minimums | 85% | 82.4% | — | 94.17% | 40.8% | 70% | |
| Meets all minimums | 84.6% | 84.5% | — | — | 23.9% | 71.3% | |
| Missing required score | 84.4% | — | — | — | — | — | |
| Meets all minimums | 84.4% | 83.5% | — | 96.67% | 23.1% | 74.4% | |
| Meets all minimums | 84.3% | 85.7% | — | — | 24.8% | 73.8% | |
| Meets all minimums | 84% | 71.5% | — | — | — | — | |
| Meets all minimums | 83.8% | 79.1% | — | — | 13.6% | — | |
| Meets all minimums | 83.73% | 79.23% | — | 90% | 18.26% | 53.73% | |
| Meets all minimums | 83.73% | 79.23% | — | — | 18.26% | 53.73% | |
| Meets all minimums | 82.5% | 81.7% | — | 92.5% | — | — | |
| Missing required score | 82% | — | — | — | 12.5% | 69.4% | |
| Missing required score | 81.2% | — | — | — | — | — | |
| Meets all minimums | 81.02% | 74.43% | — | — | — | — | |
| Missing required score | 80.6% | — | — | — | — | — | |
| Meets all minimums | 79.8% | 76.1% | — | — | 17.7% | — | |
| Meets all minimums | 79.1% | 76.2% | — | — | — | — | |
| Missing required score | 78.3% | — | — | — | 15.5% | — | |
| Missing required score | 78.29% | — | — | — | — | — | |
| Missing required score | 78.1% | — | — | — | 10.2% | — | |
| Missing required score | 75.2% | — | — | — | — | — | |
| Missing required score | 75.2% | — | — | — | — | — | |
| Meets all minimums | 74% | 65.8% | — | 82.5% | — | — | |
| Missing required score | 72.1% | — | — | — | 11.1% | — | |
| Meets all minimums | 69.6% | 62% | — | — | — | — | |
| Missing required score | 64.4% | — | 89.3% | — | — | — | |
| Below a minimum | 55.3% | — | — | — | — | — | |
| Below a minimum | 48.3% | 30.4% | 84.5% | — | — | — | |
| Below a minimum | 29.7% | 11.9% | — | — | — | — | |
| Missing required score | — | — | 79.2% | — | — | — | |
| Missing required score | — | — | — | — | — | 53.9% | |
| Missing required score | — | — | — | — | — | 62.4% | |
| Missing required score | — | — | — | — | — | 66% | |
| Missing required score | — | — | — | — | 9.92% | — | |
| Missing required score | — | — | 86% | — | — | — | |
| Missing required score | — | — | 79.6% | — | — | — | |
| Below a minimum | — | 38.89% | — | — | — | — | |
| Missing required score | — | — | — | 82.5% | — | — | |
| Missing required score | — | 80.5% | — | — | 25.2% | — | |
| Missing required score | — | — | 96.8% | — | — | — | |
| Missing required score | — | — | 91% | — | — | — | |
| Missing required score | — | — | 85.7% | — | — | — | |
| Missing required score | — | — | 86.2% | — | — | — | |
| Missing required score | — | 85.2% | — | — | 19.4% | 75.8% | |
| Missing required score | — | — | — | — | 39.2% | — | |
| Missing required score | — | — | — | — | 31% | — | |
| Missing required score | — | 71.2% | — | — | — | — | |
| Missing required score | — | — | — | — | — | 74.8% | |
| Missing required score | — | 83.8% | — | — | 12.6% | — | |
| Missing required score | — | — | — | — | 37.1% | — | |
| Missing required score | — | 67.1% | — | — | 5.2% | 47.9% | |
| Missing required score | — | 56.8% | — | — | 4.2% | 37.4% | |
| Missing required score | — | — | — | — | 19.1% | — | |
| Missing required score | — | — | 89.5% | — | — | — | |
| Missing required score | — | — | 79.9% | — | — | — | |
| Missing required score | — | — | — | 87.5% | — | — | |
| Missing required score | — | — | — | — | — | 70.6% | |
| Missing required score | — | — | 50.57% | — | — | — | |
| Missing required score | — | — | — | — | — | 52.6% | |
| Missing required score | — | — | — | — | — | 42.2% | |
| Missing required score | — | — | — | — | 22.1% | — | |
| Missing required score | — | 75.2% | — | — | 14.4% | 59.2% | |
| Missing required score | — | 86% | — | 95.83% | 30.5% | 72.8% |
EVIDENCE & REUSE
Know what the numbers say.
Scores are imported from OpenEvals leaderboard-data, whose dataset card declares an MIT license. This page includes score values and model labels, with attribution and a pinned source snapshot.
Source revision d22dd54c048a · retrieved 2026-10-08. Snapshot updates are reviewed, not live.Read this as a shortlist, not a universal ranking.
- The source aggregates reported results. Per-result verification, run settings, tool use and exact model revisions are not included.
- Each benchmark is a separate measure. We do not create an overall score or treat missing results as zero.
- We publish no benchmark questions, answers, model outputs or provider logos. The dataset’s license declaration applies to this score collection; model and underlying test materials have their own terms.
What each benchmark measures
GPQA Diamond
Graduate-level science questions.
Reported as GPQA Diamond by the source. Run settings are not included.GSM8K
Grade-school mathematical reasoning.
This older benchmark has limited coverage in the snapshot.AIME 2026
Competition-level mathematical problems.
Sampling, reasoning budget and aggregation settings are not included.Humanity’s Last Exam
Difficult questions across specialist subjects.
Tool use and text-only versus multimodal settings are not identified in this snapshot.SWE-bench Verified
Resolving real software issues.
This measures a model with an agent setup. Agent tools and scaffolding can substantially change the result.