Kimi VL A3B Thinking vs Qwen2.5 VL 32B Instruct
Wins 0 of 2 areas
—
Wins 2 of 2 areas
Reasoning · Images and charts
Qwen2.5 VL 32B Instruct is the stronger all-rounder.
Scores updated · 11 tests both models report · How we compare
Where each one wins
Tests won in each of the two areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Reasoning rests on a single test.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Qwen2.5 VL 32B Instruct pulls ahead
- College exam questions with charts, maps and diagramsMMMU+8.3points ahead
- Harder college exam questions with imagesMMMU-Pro+6.3points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+3.7points ahead
Where Kimi VL A3B Thinking pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 11 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningQwen2.5 VL 32B Instruct
- GPQA DiamondQwen2.5 VL 32B Instruct by 3.742.346+3.7
Images and chartsQwen2.5 VL 32B Instruct
Other results8 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- ScreenSpot-Pro (Acc)Kimi VL A3B Thinking by 13.452.839.4+13.4
- VideoMME (w sub.)Qwen2.5 VL 32B Instruct by 11.96677.9+11.9
- MATHKimi VL A3B Thinking by 9.691.882.2+9.6
- OSWorld-G (Acc)Kimi VL A3B Thinking by 652.546.5+6
- MMStar (Acc)Qwen2.5 VL 32B Instruct by 5.364.269.5+5.3
- MMLUKimi VL A3B Thinking by 3.68278.4+3.6
- MATH-VisionQwen2.5 VL 32B Instruct by 1.636.838.4+1.6
- MMLU-Protie68.568.8tie
Questions people ask
Which is better, Kimi VL A3B Thinking or Qwen2.5 VL 32B Instruct?
Qwen2.5 VL 32B Instruct wins all two areas where both have results: reasoning and images and charts. Kimi VL A3B Thinking wins none.
How do you compare the two?
We use the 11 benchmark tests both models have published scores on. The verdict counts the 3 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 8 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.