Skip to main content
BenchLM

MathVision

We show this table for reference; we do not rank on it.

Data verified 34 confirmed releases in the last 30 daysFollow model changes

A visual mathematics benchmark that tests whether a model can solve math problems grounded in diagrams, equations, figures, and other visual inputs.

Benchmark score on MathVision — September 27, 2026

We compile the MathVision rows from provider self-reports and secondary reports. Qwen3.8 Max leads the table at 95.2%, followed by Kimi K3 (94.3%) and Seed 2.1 Pro (92.6%). We do not use these results to rank models overall.

1Open weights

Qwen3.8 Max

Alibaba

95.2%
Overall 71.8Context 1M
2Proprietary

Kimi K3

Moonshot AI

94.3%
Overall 71.87Context 1.05M
3Proprietary

Seed 2.1 Pro

ByteDance

92.6%
Context 256K

19 modelsMultimodal & GroundedCurrentDisplay onlyUpdated September 27, 2026

Benchmark score table (19 models)

Score
1
Qwen3.8 MaxAlibaba · Open weight
95.2%
2
Kimi K3Moonshot AI · Closed
94.3%
3
Seed 2.1 ProByteDance · Closed
92.6%
4
Qwen3.8-Omni-FlashAlibaba · Closed
91.8%
5
Qwen3.8-Flash-NextAlibaba · Open weight
90.6%
6
Qwen3.7 PlusAlibaba · Closed
90.3%
7
Seed 2.1 TurboByteDance · Closed
90.1%
8
Qwen3.8-27BAlibaba · Open weight
90.0%
9
Qwen3.5 397BAlibaba · Open weight
88.6%
10
Qwen3.6 PlusAlibaba · Closed
88.0%
11
dots3-note PreviewDots Studio · Open weight
87.7%
12
Kimi K2.6Moonshot AI · Open weight
87.4%
13
Gemini 3 ProGoogle · Closed
86.6%
14
Qwen3.5-122B-A10BAlibaba · Open weight
86.2%
15
Qwen3.5-27BAlibaba · Open weight
86.0%
16
Qwen3.5-35B-A3BAlibaba · Open weight
83.9%
17
GPT-5.2OpenAI · Closed
83.0%
18
Gemma 4 12BGoogle · Open weight
79.7%
19
Claude Opus 4.5Anthropic · Closed
74.3%

Among the reported MathVision rows, Qwen3.8 Max is first at 95.2%. The third row is 2.6 points behind. The broader top-10 range is 7.2 points, so many of the published results sit in a relatively narrow band.

19 models have been evaluated on MathVision. The benchmark falls in the Multimodal & Grounded category. MathVision is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.

About MathVision

Year

2026

Tasks

Visually grounded math problems

Format

Image + math reasoning

Difficulty

Advanced multimodal mathematics

MathVision matters because text-only math ability does not guarantee strong performance when the relevant information is embedded in images, geometry diagrams, or formatted equations.

Qwen3.6 launch benchmarks

Freshness and provenance

Version

MathVision 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.

Questions

What does MathVision measure?

A visual mathematics benchmark that tests whether a model can solve math problems grounded in diagrams, equations, figures, and other visual inputs.

Which model scores highest on MathVision?

Qwen3.8 Max by Alibaba currently leads with a score of 95.2% on MathVision.

How many models are evaluated on MathVision?

19 AI models have been evaluated on MathVision on BenchLM.

Compare top models on MathVision

Qwen3.8 Max vs Kimi K3Kimi K3 vs Seed 2.1 ProSeed 2.1 Pro vs Qwen3.8-Omni-FlashQwen3.8-Omni-Flash vs Qwen3.8-Flash-Next
Last updated: September 27, 2026 · BenchLM version MathVision 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.