BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases7

MMMU-Pro (Vals) leaderboard

As of 6 Nov 2026, the top score on MMMU-Pro (Vals) is 90.6% by Claude Fable 5.1 (Anthropic), ahead of Claude Opus 5 at 89.9%, across 91 model configurations with a published result (Vals AI, result dated 1 Sep 2026). Vals AI's run of the harder ten-option MMMU-Pro multimodal exam. MMMU-Pro (Vals) is shown for reference only and does not count toward the BenchLeader Index.

Key facts (as of 6 Nov 2026)

  • Claude Fable 5.1 holds the top MMMU-Pro (Vals) score at 90.6% (Vals AI, 1 Sep 2026).
  • Claude Opus 5 is second on MMMU-Pro (Vals) at 89.9% (Vals AI, 1 Sep 2026).
  • MMMU-Pro (Vals): Vals AI's own run of MMMU-Pro. Run by Vals AI (Vals AI).
  • 91 model configurations have a published MMMU-Pro (Vals) result on BenchLeader (BenchLeader, 6 Nov 2026).
  • MMMU-Pro (Vals) does not feed the BenchLeader Index; it is shown for reference (BenchLeader methodology).

Vals AI's own run of MMMU-Pro. Run by Vals AI.

Published by
Vals AI
Category
Multimodal
Index weight
Reference only
Models
91
Data as of
6 Nov 2026

Vals AI (vals.ai).

What the test looks like

Vals AI's run of the harder ten-option MMMU-Pro multimodal exam.

How it is scored

Accuracy, run by Vals AI.

What to keep in mind

Shown next to the Artificial Analysis run and the official board.

Top 15
  1. 1Claude Fable 5.190.6%
  2. 2Claude Opus 589.9%
  3. 3Claude Fable 589.3%
  4. 4Gemini 3.8 Flash (high)89.1%
  5. 5Gemini 3.7 Flash (high)89.0%
  6. 6GPT-5.6 Sol (max)88.8%
  7. 7Gemini 3.6 Flash (high)88.4%
  8. 8GPT-5.5 (xhigh)88.3%
  9. 9Gemini 3.5 Flash (high)88.3%
  10. 10Gemini 3.1 Pro (high)88.2%
  11. 11Kimi K388.2%
  12. 12Qwen3.8 Max (max)88.0%
  13. 13Gemini 3 Flash (high)87.6%
  14. 14GPT-5.4 (xhigh)87.5%
  15. 15Gemini 3 Pro (high)87.5%
91 of 91
#
1Claude Fable 5.190.6%66.4$0.1662026-09-01
2Claude Opus 589.9%65.2$0.0692026-09-01
3Claude Fable 589.3%66.5$0.1612026-09-01
4Gemini 3.8 Flashhigh89.1%62.0$0.00452026-09-01
5Gemini 3.7 Flashhigh89.0%62.2$0.0142026-09-01
6GPT-5.6 Solmax88.8%65.4$0.0562026-09-01
7Gemini 3.6 Flashhigh88.4%61.2$0.0192026-09-01
8GPT-5.5xhigh88.3%64.3$0.0432026-09-01
9Gemini 3.5 Flashhigh88.3%62.5$0.0252026-09-01
10Gemini 3.1 Prohigh88.2%60.1$0.0642026-09-01
11Kimi K388.2%59.1$0.0592026-09-01
12Qwen3.8 Maxmax88.0%58.8$0.0162026-09-01
13Gemini 3 Flashhigh87.6%58.3$0.0142026-09-01
14GPT-5.4xhigh87.5%63.2$0.0472026-09-01
15Gemini 3 Prohigh87.5%60.1$0.0102026-09-01
16Muse Spark87.4%64.1$0.00172026-09-01
17GPT-5.2xhigh86.7%60.0$0.0612026-09-01
18Claude Opus 4.886.6%60.0$0.1212026-09-01
19Muse Spark 1.1xhigh86.6%59.5$0.0142026-09-01
20GPT-5.6 Terramax86.5%60.7$0.0162026-09-01
21Kimi K2.686.3%55.2$0.0352026-09-01
22Muse Spark 1.2xhigh86.1%61.7$0.0182026-09-01
23GLM 5.3 Flashmax86.0%54.2$0.00102026-09-01
24Claude Opus 4.785.5%63.1$0.1702026-09-01
25GPT-5.6 Lunamax85.0%56.3$0.00642026-09-01
26Kimi K2.5thinking84.3%58.7$0.0202026-09-01
27Qwen3.6 Plus84.2%59.1$0.0172026-09-01
28Qwen3.8 27Bxhigh83.9%52.5$0.0132026-09-01
29Claude Opus 4.6thinking83.9%62.4$0.1162026-09-01
30Claude Sonnet 4.683.6%58.2$0.1442026-09-01
31Gemini 3.5 Flash Litehigh83.6%47.9$0.00662026-09-01
32Grok 4.20thinking83.5%58.4$0.0272026-09-01
33GPT-5.1high83.2%57.7$0.00282026-09-01
34Grok 4.383.1%54.5$0.0222026-09-01
35Claude Sonnet 583.0%55.9$0.0332026-09-01
36Claude Opus 4.5thinking83.0%59.6$0.1272026-09-01
37Gemini 3.1 Flash Litehigh82.5%48.5$0.00322026-09-01
38Qwen3.5 Flash81.9%51.8$0.00262026-09-01
39GPT-5high81.5%57.9$0.0332026-09-01
40Gemini 2.5 Pro81.3%54.5$0.00782026-09-01
41MiniMax M381.2%56.4$0.00942026-09-01
42Claude Opus 4.581.1%57.9$0.0222026-09-01
43Gemini 2.5 Flash 09 2025thinking80.8%52.8$0.00192026-09-01
44o3high80.4%53.8$0.0152026-09-01
45o4-minihigh79.7%50.3$0.0112026-09-01
46Gemini 2.5 Flash 09 202579.5%46.8$0.00192026-09-01
47Claude Sonnet 4.5thinking79.3%52.2$0.0582026-09-01
48GPT-5.4 Minixhigh79.3%56.0$0.00712026-09-01
49GPT-5 minihigh78.9%54.6$0.00722026-09-01
50Inkling78.5%47.5$0.0312026-09-01
51Claude Opus 4.1thinking77.5%57.8$0.1692026-09-01
52o1high77.4%48.3$0.1462026-09-01
53Grok 476.3%58.3$0.0982026-09-01
54Gemini 2.5 Flash Lite 09thinking75.4%49.7$0.00042026-09-01
55Claude 3.7 Sonnetthinking75.1%51.9$0.0512026-09-01
56Claude Sonnet 4thinking74.9%53.3$0.0512026-09-01
57Claude Opus 4.173.7%51.4$0.0442026-09-01
58GPT-5.4 Nanohigh73.6%48.3$0.00082026-09-01
59Claude Opus 473.3%52.6$0.0622026-09-01
60Grok 4 Fastthinking72.8%51.6$0.00292026-09-01

More multimodal benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on MMMU-Pro (Vals)

Cite as: BenchLeader, “MMMU-Pro (Vals) leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/vals_mmmu_pro.

MMMU-Pro (Vals): questions

What does MMMU-Pro (Vals) measure?
Vals AI's run of the harder ten-option MMMU-Pro multimodal exam. Scores are reported in percent of tasks solved; higher is better.
What is the top score on MMMU-Pro (Vals)?
As of 6 Nov 2026, the highest score on MMMU-Pro (Vals) is 90.6% by Claude Fable 5.1 (result dated 1 Sep 2026), ahead of Claude Opus 5 at 89.9%. Scores come from Vals AI and are re-read daily.
Which AI model leads MMMU-Pro (Vals)?
Claude Fable 5.1 leads MMMU-Pro (Vals) with 90.6% as of 6 Nov 2026 (result dated 1 Sep 2026), ahead of Claude Opus 5 at 89.9%.
How many models have MMMU-Pro (Vals) results?
91 model configurations have a MMMU-Pro (Vals) result on BenchLeader, all taken from Vals AI.
Who runs MMMU-Pro (Vals) and how often is it updated?
MMMU-Pro (Vals) is published by Vals AI. BenchLeader re-reads the published results once a day and records the date each result was published.
Does MMMU-Pro (Vals) count toward the BenchLeader Index?
No. MMMU-Pro (Vals) is shown for reference but left out of the composite index.