MetaAuto-detectedLlama 4 Maverick Basic
Best configuration ranks #542 of 610 on the BenchLeader Index at 38.8 ±10.7. Last measured 1 Sept 2026.
- Blended price
- –
- Output speed
- –
- First answer
- –
- Context
- –
- Overall index39
- Coding31
- Maths44
- Knowledge31
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | default | Source |
|---|---|---|
| GPQA Diamond (Vals)not in index | 69.4%#92 | Vals AI |
Coding
| Benchmark | default | Source |
|---|---|---|
| LiveCodeBench | 47.3%#113 | Vals AI |
Maths
| Benchmark | default | Source |
|---|---|---|
| AIME (Vals) | 25.2%#76 | Vals AI |
| MGSM | 92.4%#18 | Vals AI |
Knowledge
| Benchmark | default | Source |
|---|---|---|
| MMLU-Pro | 79.4%#92 | Vals AI |
| LegalBench | 77.8%#97 | Vals AI |
| CorpFin | 49.7%#104 | Vals AI |
| TaxEval | 66.6%#107 | Vals AI |
| MedQA | 43.3%#89 | Vals AI |
| PRBench Finance | 22.4%#33 | Scale AI SEAL |
| PRBench Legal | 24.8%#33 | Scale AI SEAL |
See also
Data as of 9 Sept 2026. Compare with another model.