VideoMMMU (VideoMMMU)

A video extension of MMMU-style multimodal reasoning over expert questions grounded in temporal media.

Top Models on VideoMMMU — March 2026

As of March 2026, Gemini 3 Pro leads the VideoMMMU leaderboard with 87.6% , followed by Kimi K2.5 (86.6%) and GPT-5.2 (85.9%).

11 modelsMultimodal & GroundedCurrentDisplay onlyUpdated April 2, 2026

According to BenchLM.ai, Gemini 3 Pro leads the VideoMMMU benchmark with a score of 87.6%, followed by Kimi K2.5 (86.6%) and GPT-5.2 (85.9%). The top models are clustered within 1.7 points, suggesting this benchmark is nearing saturation for frontier models.

11 models have been evaluated on VideoMMMU. The benchmark falls in the Multimodal & Grounded category. This category carries a 12% weight in BenchLM.ai's overall scoring system. VideoMMMU is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About VideoMMMU

Year

2026

Tasks

Video-grounded expert reasoning

Format

Video + text reasoning

Difficulty

Frontier multimodal video reasoning

VideoMMMU tests whether multimodal reasoning skills extend from static images into temporal video understanding. It is useful for evaluating long-form visual reasoning rather than static scene recognition.

Qwen3.6 launch benchmarks

BenchLM freshness & provenance

Version

VideoMMMU 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

Leaderboard (11 models)

#1Gemini 3 Pro
87.6%
#2Kimi K2.5
86.6%
#3GPT-5.2
85.9%
#4Qwen3.5 397B
84.7%
#5Claude Opus 4.5
84.4%
#6Qwen3.6 Plus
84.0%
#7Gemini 2.5 Pro
83.6%
#8o3
83.3%
#9Claude 3.5 Sonnet
65.8%
#10GPT-4o
61.2%
#11Gemini 1.5 Pro
53.9%

FAQ

What does VideoMMMU measure?

A video extension of MMMU-style multimodal reasoning over expert questions grounded in temporal media.

Which model scores highest on VideoMMMU?

Gemini 3 Pro by Google currently leads with a score of 87.6% on VideoMMMU.

How many models are evaluated on VideoMMMU?

11 AI models have been evaluated on VideoMMMU on BenchLM.

Last updated: April 2, 2026 · BenchLM version VideoMMMU 2026

Weekly LLM Benchmark Digest

Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.

Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.