Early coverage — no capability classification yet clears Limited. Insufficient data on 10 of 10 axes.
| Benchmark | Score | Settings | Source | Verification | Methodology | Measured | Reference |
|---|---|---|---|---|---|---|---|
DocVQA_VAL | 92.04 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
TextVQA_VAL | 82.15 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
OCRVQA_TEST | 39.03 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
AI2D_TEST | 85.07 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
| 90.59 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ | |
Q-Bench1_VAL | 75.99 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
| 71.11 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ | |
| 68.40 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ | |
| 55.87 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ | |
MTVQA_TEST | 30.30 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
MMT-Bench_VAL | 65.17 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
ScienceQA_VAL | 97.81 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
ScienceQA_TEST | 97.97 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
MMMU_VAL | 56.11 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
MMVet | 80.28 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
SEEDBench2_Plus | 69.52 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
SEEDBench_IMG | 77.00 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
CCBench | 77.84 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
A-Bench_VAL | 75.86 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
InHouse Dataset B | 36.25 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
InHouse Dataset A | 40.64 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
| 49.20 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ | |
ChartQA Pro | 37.32 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
Mathverse | 43.68 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
Mathvision | 29.61 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |
Mathvista-mini | 69.50 | — | huggingface.coREADME | UNVERIFIED | VENDOR | 1h ago | view ↗ |