benchmark evidence
MMMU-Pro
MMMU-Pro: vision-only inputs prevent text shortcuts — harder multimodal reasoning across academic disciplines.
activeupstream source ->
winner on MMMU-Pro
OpenAI: GPT-5.583.2
direct benchmark result, not a broad vertical composite | source row dated 2000-01-01
scored on 2000-01-01 · stale source data (9650d)
latest mapped results | top 20
| # | Model | Score | Evidence | Tested |
|---|---|---|---|---|
| 1 | OpenAI: GPT-5.5 | 83.2 | model-only independent_benchmark | 2000-01-01 |
| 2 | OpenAI: GPT-5.4 | 81.2 | model-only independent_benchmark | 2000-01-01 |
| 3 | OpenAI: GPT-5.2 | 79.5 | model-only independent_benchmark | 2000-01-01 |
| 4 | OpenAI: GPT-5 | 78.4 | model-only independent_benchmark | 2000-01-01 |
| 5 | Anthropic: Claude Opus 4.6 | 77.3 | model-only independent_benchmark | 2000-01-01 |
| 6 | OpenAI: GPT-5.4 Mini | 76.6 | model-only independent_benchmark | 2000-01-01 |
| 7 | OpenAI: o3 | 76.4 | model-only independent_benchmark | 2000-01-01 |
| 8 | Anthropic: Claude Sonnet 4.6 | 75.6 | model-only independent_benchmark | 2000-01-01 |
| 9 | Meta: Llama 4 Maverick | 59.6 | model-only independent_benchmark | 2000-01-01 |
what this result means
MMMU-Pro: vision-only inputs prevent text shortcuts — harder multimodal reasoning across academic disciplines.
This benchmark contributes direct public evidence. Read its scope before generalizing the result.
A win here is a win on MMMU-Pro. Broad task pages require independent corroboration before naming a general winner.
source record
category: multimodal
metric: accuracy
matched models: 9
latest source date: 2000-01-01
direction: higher is better