AI capability, measured with evidence
QBench Index 1.2 compares models across five source-audited evaluations grouped into four practical task families. Every result keeps its original metric, configuration and source; missing scores are never guessed.
QBench Index 1.2 leaderboard
How QBench Index 1.2 works
Each eligible benchmark is converted to a midrank percentile within its own evaluated model cohort. Results are averaged inside four QBench task families, then averaged equally. The final score is adjusted toward 50 when a model has fewer than four task families of evidence.
Rank: at least 2 eligible evaluations across 2+ task families · Evidence: full (4 families), broad (3), or limited (2) · Insufficient evidence: retained in Models, without an Index score.
Pareto frontier: open “Index vs. API price + Pareto”. The dashed frontier is on by default; the API token-price proxy is not cost per task.
Models
Models with at least one eligible result. Routers and specialist systems are identified separately. The API price catalog below also includes priced models outside the composite Index.
API price catalog
Per 1M input/output tokens. Official provider API and OpenRouter route prices are labelled separately.
| Model | Provider | Input / 1M | Output / 1M | Price source | Conditions |
|---|
Comparison
Category radar uses the index percentiles. Empty spokes mean the category has no result for that model.
Benchmark directory
Select a benchmark to see its model scores. The status shows whether it enters the index.