Benchmarks · knowledge
MMMU Pro (Vals)
Multimodal Multi-task Benchmark Scores are the values published by VALS AI. Source benchmark version 1.
32 models · 32 results · updated 2026-09-01 · vals.ai ↗
Leaderboard
Fable 5.1 max
Heavy line: best score at each price. Thin lines join one model's effort levels. Cost uses each model's median subscription plan, or API price with no plan data. 19 more models under All models in settings.
More from this benchmark
1 published measuresTime per task93 results · lower is better
19 more models under All models in settings.
All results
32 rows| # | Model | Effort | MMMU Pro (Vals) | Subscription $ / task | API $ / task | Time |
|---|---|---|---|---|---|---|
| 1 | Fable 5.1 | max | 90.6% | $0.026 | $0.17 | 37s |
| 2 | Opus 5 | max | 89.9% | $0.001 | $0.069 | 30s |
| 3 | Fable 5 | max | 89.3% | $0.018 | $0.16 | 61s |
| 4 | Gemini 3.8 Flash | high | 89.1% | $0.000 | $0.004 | 12s |
| 5 | Gemini 3.7 Flash | high | 89.0% | $0.002 | $0.014 | 6.3s |
| 6 | GPT-5.6 Sol | max | 88.8% | $0.002 | $0.056 | 26s |
| 7 | Gemini 3.6 Flash | high | 88.4% | $0.001 | $0.019 | 11s |
| 8 | Gemini 3.5 Flash | high | 88.3% | $0.025 | $0.025 | 12s |
| 9 | GPT-5.5 | xhigh | 88.3% | $0.001 | $0.043 | 54s |
| 10 | Kimi K3 | max | 88.2% | $0.005 | $0.059 | 1.8 min |
| 11 | Qwen 3.8 Max | max | 88.0% | $0.002 | $0.016 | 44s |
| 12 | Muse Spark | 87.4% | $0.002 | $0.002 | 48s | |
| 13 | Opus 4.8 | max | 86.6% | $0.002 | $0.12 | 49s |
| 14 | Muse Spark 1.1 | xhigh | 86.6% | $0.014 | $0.014 | 32s |
| 15 | GPT-5.6 Terra | max | 86.5% | $0.000 | $0.016 | 13s |
| 16 | Kimi K2.6 | max | 86.3% | $0.017 | $0.035 | 4.0 min |
| 17 | Muse Spark 1.2 | xhigh | 86.1% | $0.009 | $0.018 | 74s |
| 18 | GLM 5.3 Flash | max | 86.0% | free | $0.001 | 28s |
| 19 | Opus 4.7 | max | 85.5% | $0.17 | $0.17 | 71s |
| 20 | GPT-5.6 Luna | max | 85.0% | $0.000 | $0.006 | 35s |
| 21 | Qwen 3.6 Plus | max | 84.2% | $0.009 | $0.017 | 1.9 min |
| 22 | Qwen 3.8 27B | xhigh | 83.9% | $0.001 | $0.013 | 43s |
| 23 | Sonnet 4.6 | max | 83.6% | $0.14 | $0.14 | 2.2 min |
| 24 | Gemini 3.5 Flash-Lite | high | 83.6% | $0.007 | $0.007 | 7.8s |
| 25 | Grok 4.3 | xhigh | 83.1% | $0.022 | $0.022 | 2.5 min |
| 26 | Sonnet 5 | max | 83.0% | $0.001 | $0.033 | 19s |
| 27 | Qwen 3.5 Flash | max | 81.9% | $0.003 | $0.003 | 46s |
| 28 | MiniMax M3 | max | 81.2% | $0.002 | $0.009 | 68s |
| 29 | MiMo V2.5 | xhigh | 80.0% | $0.001 | $0.001 | 42s |
| 30 | Inkling | max | 78.5% | $0.003 | $0.031 | 2.1 min |