DynaMath (DynaMath)

A multimodal benchmark for dynamic mathematical reasoning over visual and structured inputs.

Top Models on DynaMath — March 2026

As of March 2026, Qwen3.6 Plus leads the DynaMath leaderboard with 88.0% , followed by GPT-5.2 (86.8%) and Qwen3.5 397B (86.3%).

6 modelsMultimodal & GroundedCurrentDisplay onlyUpdated April 2, 2026

According to BenchLM.ai, Qwen3.6 Plus leads the DynaMath benchmark with a score of 88.0%, followed by GPT-5.2 (86.8%) and Qwen3.5 397B (86.3%). The top models are clustered within 1.7 points, suggesting this benchmark is nearing saturation for frontier models.

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

About DynaMath

Year

2026

Tasks

Dynamic visual math problems

Format

Multimodal mathematical reasoning

Difficulty

Advanced multimodal mathematics

DynaMath is intended to probe whether models can track changing mathematical structure in multimodal settings rather than solving static text-only equations.

Qwen3.6 launch benchmarks

BenchLM freshness & provenance

Version

DynaMath 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)

#1Qwen3.6 Plus
88.0%
#2GPT-5.2
86.8%
#3Qwen3.5 397B
86.3%
#4Gemini 3 Pro
85.1%
#5Kimi K2.5
84.4%
#6Claude Opus 4.5
79.7%

FAQ

What does DynaMath measure?

A multimodal benchmark for dynamic mathematical reasoning over visual and structured inputs.

Which model scores highest on DynaMath?

Qwen3.6 Plus by Alibaba currently leads with a score of 88.0% on DynaMath.

How many models are evaluated on DynaMath?

6 AI models have been evaluated on DynaMath on BenchLM.

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