MStar (MStar)

A general visual question-answering benchmark used in provider tables for real-image reasoning quality.

Top Models on MStar — March 2026

As of March 2026, Qwen3.5 397B leads the MStar leaderboard with 83.8% , followed by Qwen3.6 Plus (83.3%) and Gemini 3 Pro (83.1%).

6 modelsMultimodal & GroundedCurrentDisplay onlyUpdated April 2, 2026

According to BenchLM.ai, Qwen3.5 397B leads the MStar benchmark with a score of 83.8%, followed by Qwen3.6 Plus (83.3%) and Gemini 3 Pro (83.1%). The top models are clustered within 0.7 points, suggesting this benchmark is nearing saturation for frontier models.

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

About MStar

Year

2026

Tasks

Real-image visual QA

Format

Image-grounded QA

Difficulty

General visual reasoning

MStar sits between broad multimodal reasoning and grounded VQA. It is useful for checking whether a model can answer real-image questions without the stronger domain structure of office or academic benchmarks.

Qwen3.6 launch benchmarks

BenchLM freshness & provenance

Version

MStar 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 (6 models)

#1Qwen3.5 397B
83.8%
#2Qwen3.6 Plus
83.3%
#3Gemini 3 Pro
83.1%
#4Kimi K2.5
80.5%
#5GPT-5.2
77.1%
#6Claude Opus 4.5
73.2%

FAQ

What does MStar measure?

A general visual question-answering benchmark used in provider tables for real-image reasoning quality.

Which model scores highest on MStar?

Qwen3.5 397B by Alibaba currently leads with a score of 83.8% on MStar.

How many models are evaluated on MStar?

6 AI models have been evaluated on MStar on BenchLM.

Last updated: April 2, 2026 · BenchLM version MStar 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.