MathVision with Python (MathVision w/ Python)
We show this table for reference; we do not rank on it.
A tool-augmented MathVision variant that permits Python during visual mathematics reasoning.
Benchmark score on MathVision w/ Python — November 13, 2026
We compile the MathVision w/ Python rows from provider self-reports. Kimi K3 leads the table at 97.8%, followed by Qwen3.8 Max (97.7%) and LongCat-2.5 (96.5%). We do not use these results to rank models overall.
Kimi K3
Moonshot AI
Reported by Moonshot AIQwen3.8 Max
Alibaba
Reported by AlibabaLongCat-2.5
Meituan
Reported by Meituan6 modelsMultimodal & GroundedCurrentDisplay onlyUpdated November 13, 2026
Benchmark score table (6 models)
All 6 results on this page were reported by the labs that built the models.
| Rank | Model / configuration | Score | Who ran it | Parameters (B) | Open / closed |
|---|---|---|---|---|---|
| 1 | Kimi K3Moonshot AIReported by Moonshot AI | 97.8% | Reported by Moonshot AI | Not reported | Open |
| 2 | Qwen3.8 MaxAlibabaReported by Alibaba | 97.7% | Reported by Alibaba | Not reported | Open |
| 3 | LongCat-2.5MeituanReported by Meituan | 96.5% | Reported by Meituan | Not reported | Open |
| 4 | Qwen3.8-Omni-FlashAlibabaReported by Alibaba | 96.2% | Reported by Alibaba | Not reported | Closed |
| 5 | Qwen3.8-Flash-NextAlibabaReported by Alibaba | 95.7% | Reported by Alibaba | Not reported | Open |
| 6 | Qwen3.8-27BAlibabaReported by Alibaba | 94.6% | Reported by Alibaba | Not reported | Open |
- Lab-reported
- Benchmark-run
- Independent
Among the reported MathVision w/ Python rows, Kimi K3 is first at 97.8%. The third row is 1.3 points behind. The broader top-10 range is 3.2 points, so many of the published results sit in a relatively narrow band.
6 models have been evaluated on MathVision w/ Python. The benchmark falls in the Multimodal & Grounded category. MathVision w/ Python is currently displayed for reference but excluded from the scoring formula, so it does not directly affect overall rankings.
About MathVision w/ Python
Year
2026
Tasks
Visual mathematics problems with Python
Format
Image and mathematics reasoning with tools
Difficulty
Advanced multimodal mathematics
BenchLM stores the Python-assisted value separately from the tool-free MathVision key.
Freshness and provenance
Version
MathVision w/ Python 2026
Refresh cadence
Quarterly
Staleness state
Current
Question availability
Public benchmark set
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 w/ Python measure?
A tool-augmented MathVision variant that permits Python during visual mathematics reasoning.
Which model scores highest on MathVision w/ Python?
Kimi K3 by Moonshot AI currently leads with a score of 97.8% on MathVision w/ Python.
How many models are evaluated on MathVision w/ Python?
6 AI models have published results on MathVision w/ Python in the BenchLM catalog.
Compare top models on MathVision w/ Python
Know when it’s worth switching models
The model to choose, the cheaper alternative, and the release we would wait on.
Read a sample issueJoin 7,000+ subscribers.
One email each week. Unsubscribe anytime.