Qwen3 VL 4B (Reasoning) is the stronger all-rounder.
Scores updated · 25 tests both models report · How we compare
Where each one wins
Tests won in each of the one area 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.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Every test, side by side
All 25 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Images and chartsQwen3 VL 4B (Reasoning)
Other results21 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.
- VSI-BenchQwen3 VL 4B (Reasoning) by 29.725.555.2+29.7
- MultiDocVQA-valQwen3 VL 4B (Reasoning) by 28.858.487.2+28.8
- InfoVQA (val)Qwen3 VL 4B (Reasoning) by 24.155.479.5+24.1
- MV-BenchQwen3 VL 4B (Reasoning) by 20.149.269.3+20.1
- LVBenchQwen3 VL 4B (Reasoning) by 19.134.453.5+19.1
- DocVQA-valQwen3 VL 4B (Reasoning) by 18.576.294.7+18.5
- MME-PerceptionQwen3 VL 4B (Reasoning) by 114 rating points15901704+114 rating
- ChartQAProQwen3 VL 4B (Reasoning) by 10.825.436.2+10.8
- NextQAQwen3 VL 4B (Reasoning) by 10.86979.8+10.8
- SATPhi 4 R V 15B by 1066.756.7+10
- MLVU-devQwen3 VL 4B (Reasoning) by 9.751.861.5+9.7
- LongVideoBenchQwen3 VL 4B (Reasoning) by 6.551.257.7+6.5
- MME-RealWorldQwen3 VL 4B (Reasoning) by 5.457.863.2+5.4
- Timelens-ActivityNetQwen3 VL 4B (Reasoning) by 5.42328.4+5.4
- Ref-DAVIS17Qwen3 VL 4B (Reasoning) by 5.32.17.5+5.3
- TextVQA-valQwen3 VL 4B (Reasoning) by 4.576.180.5+4.5
- CV-BenchQwen3 VL 4B (Reasoning) by 4.481.385.7+4.4
- AI2D w/ MaskPhi 4 R V 15B by 1.482.981.5+1.4
- MMBench-CN-devQwen3 VL 4B (Reasoning) by 1.179.580.6+1.1
- MMBench-EN-devtie84.283.3tie
- ChartQAtie83.483.3tie
Questions people ask
Which is better, Phi 4 R V 15B or Qwen3 VL 4B (Reasoning)?
Qwen3 VL 4B (Reasoning) wins the one area where both have results: images and charts. Phi 4 R V 15B wins none.
How do you compare the two?
We use the 25 benchmark tests both models have published scores on. The verdict counts the 4 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 21 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.