Qwen3 VL 235B A22B Reasoning vs Qwen3 VL 4B Instruct
Wins 7 of 7 areas
Coding · Agents · Reasoning · Facts · Images and charts · Long documents · Following instructions
Wins 0 of 7 areas
—
Qwen3 VL 235B A22B Reasoning is the stronger all-rounder.
Scores updated · 47 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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.
- Images and chartsUnderstanding pictures, charts and video50Qwen3 VL 235B A22B Reasoning5 of 6 tests · 1 tie
- CodingWriting and fixing software30Qwen3 VL 235B A22B Reasoning3 of 3 tests
- ReasoningHard problems that need careful thinking20Qwen3 VL 235B A22B Reasoning2 of 3 tests · 1 tie
- FactsGetting facts right instead of making them up20Qwen3 VL 235B A22B Reasoning2 of 2 tests
- AgentsCarrying out multi-step tasks on its own10Qwen3 VL 235B A22B Reasoning1 of 1 test
- Long documentsFinding answers in very long texts10Qwen3 VL 235B A22B Reasoning1 of 1 test
- Following instructionsDoing exactly what it is asked10Qwen3 VL 235B A22B Reasoning1 of 1 test
Agents, long documents and following instructions rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Qwen3 VL 235B A22B Reasoning pulls ahead
- Reasons across sets of long documentsAA-LCR+49.7points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+40.1points ahead
- Recent programming contest problemsLiveCodeBench v6+32.2points ahead
Where Qwen3 VL 4B Instruct pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 47 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3 VL 235B A22B Reasoning
- LiveCodeBench v6Qwen3 VL 235B A22B Reasoning by 32.270.137.9+32.2
- SciCodeQwen3 VL 235B A22B Reasoning by 26.239.913.7+26.2
- Terminal-Bench HardQwen3 VL 235B A22B Reasoning by 11.411.40+11.4
AgentsQwen3 VL 235B A22B Reasoning
- OSWorld-VerifiedQwen3 VL 235B A22B Reasoning by 11.938.126.2+11.9
ReasoningQwen3 VL 235B A22B Reasoning
- GPQA DiamondQwen3 VL 235B A22B Reasoning by 40.177.237.1+40.1
- Humanity's Last ExamQwen3 VL 235B A22B Reasoning by 8.311.93.6+8.3
- CritPttie00tie
FactsQwen3 VL 235B A22B Reasoning
- AA-Omniscience · Non-hallucinationQwen3 VL 235B A22B Reasoning by 12.214.82.6+12.2
- AA-Omniscience · AccuracyQwen3 VL 235B A22B Reasoning by 1020.910.9+10
Images and chartsQwen3 VL 235B A22B Reasoning
- CharXiv (RQ)Qwen3 VL 235B A22B Reasoning by 26.466.139.7+26.4
- MMMU-ProQwen3 VL 235B A22B Reasoning by 24.868.743.9+24.8
- MathVistaQwen3 VL 235B A22B Reasoning by 12.185.873.7+12.1
- MMStarQwen3 VL 235B A22B Reasoning by 8.978.769.8+8.9
- BLINKQwen3 VL 235B A22B Reasoning by 1.367.165.8+1.3
- OCRBenchtie87.588.1tie
Long documentsQwen3 VL 235B A22B Reasoning
- AA-LCRQwen3 VL 235B A22B Reasoning by 49.763.714+49.7
Following instructionsQwen3 VL 235B A22B Reasoning
- IFBenchQwen3 VL 235B A22B Reasoning by 24.756.531.8+24.7
Other results30 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.
- HMMT 2025Qwen3 VL 235B A22B Reasoning by 46.777.430.7+46.7
- AIME 2025Qwen3 VL 235B A22B Reasoning by 43.189.746.6+43.1
- τ²-Bench Telecom (AA run)Qwen3 VL 235B A22B Reasoning by 30.754.123.4+30.7
- AA-OmniscienceQwen3 VL 235B A22B Reasoning by 29.5-46.5-76+29.5
- SuperGPQAQwen3 VL 235B A22B Reasoning by 2464.340.3+24
- VideoMMMUQwen3 VL 235B A22B Reasoning by 23.88056.2+23.8
- MATH-VisionQwen3 VL 235B A22B Reasoning by 2374.651.6+23
- MMLU-ProXQwen3 VL 235B A22B Reasoning by 21.280.659.4+21.2
- IncludeQwen3 VL 235B A22B Reasoning by 18.68061.4+18.6
- MMLU-ProQwen3 VL 235B A22B Reasoning by 16.783.867.1+16.7
- MMMU (val) (Pass@1)Qwen3 VL 235B A22B Reasoning by 13.280.667.4+13.2
- MMLU-ReduxQwen3 VL 235B A22B Reasoning by 12.293.781.5+12.2
- ERQAQwen3 VL 235B A22B Reasoning by 11.252.541.3+11.2
- RealWorldQAQwen3 VL 235B A22B Reasoning by 10.481.370.9+10.4
- InfoVQA_testQwen3 VL 235B A22B Reasoning by 9.289.580.3+9.2
- HallusionBenchQwen3 VL 235B A22B Reasoning by 9.166.757.6+9.1
- BFCL v3Qwen3 VL 235B A22B Reasoning by 8.671.963.3+8.6
- CharadesSTAQwen3 VL 235B A22B Reasoning by 863.555.5+8
- AA IntelligenceQwen3 VL 235B A22B Reasoning by 7.713.45.7+7.7
- LVBenchQwen3 VL 235B A22B Reasoning by 7.463.656.2+7.4
- IFEvalQwen3 VL 235B A22B Reasoning by 5.988.282.3+5.9
- OCRBench v2 (Chinese)Qwen3 VL 235B A22B Reasoning by 5.963.557.6+5.9
- CC-OCRQwen3 VL 235B A22B Reasoning by 5.381.576.2+5.3
- AI2DQwen3 VL 235B A22B Reasoning by 5.189.284.1+5.1
- SimpleQAQwen3 VL 4B Instruct by 3.644.448+3.6
- OCRBench v2_enQwen3 VL 235B A22B Reasoning by 3.166.863.7+3.1
- ScreenSpot-Pro (No tools)Qwen3 VL 235B A22B Reasoning by 2.361.859.5+2.3
- ScreenSpotQwen3 VL 235B A22B Reasoning by 1.495.494+1.4
- DocVQA_testQwen3 VL 235B A22B Reasoning by 1.296.595.3+1.2
- MM MTBenchQwen3 VL 235B A22B Reasoning by 18.57.5+1
Questions people ask
Which is better, Qwen3 VL 235B A22B Reasoning or Qwen3 VL 4B Instruct?
Qwen3 VL 235B A22B Reasoning wins all seven areas where both have results: coding, agents, reasoning, facts, images and charts, long documents and following instructions. Qwen3 VL 4B Instruct wins none.
Which is better for coding?
Qwen3 VL 235B A22B Reasoning. It wins 3 of the 3 coding tests both models report; Qwen3 VL 4B Instruct wins none.
How do you compare the two?
We use the 47 benchmark tests both models have published scores on. The verdict counts the 17 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 30 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.