Qwen3 VL 30B A3B Reasoning vs Qwen3 VL 32B Reasoning
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Coding · Reasoning · Facts · Images and charts · Long documents · Following instructions
Qwen3 VL 32B Reasoning is the stronger all-rounder.
Scores updated · 50 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 video06Qwen3 VL 32B Reasoning6 of 6 tests
- CodingWriting and fixing software02Qwen3 VL 32B Reasoning2 of 3 tests · 1 tie
- ReasoningHard problems that need careful thinking02Qwen3 VL 32B Reasoning2 of 3 tests · 1 tie
- FactsGetting facts right instead of making them up01Qwen3 VL 32B Reasoning1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts01Qwen3 VL 32B Reasoning1 of 1 test
- Following instructionsDoing exactly what it is asked01Qwen3 VL 32B Reasoning1 of 1 test
- AgentsCarrying out multi-step tasks on its own11Even1 each
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.
Qwen3 VL 32B Reasoning costs 69% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Qwen3 VL 32B Reasoning pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+14.3points ahead
- Reasons across sets of long documentsAA-LCR+12.6points ahead
- Completes tasks by operating a computer desktopOSWorld-Verified+10.4points ahead
Where Qwen3 VL 30B A3B Reasoning pulls ahead
- Real work tasks from 44 professionsGDPVal+1points ahead
Every test, side by side
All 50 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3 VL 32B Reasoning
- Terminal-Bench HardQwen3 VL 32B Reasoning by 2.35.37.6+2.3
- LiveCodeBench v6Qwen3 VL 32B Reasoning by 1.464.265.6+1.4
- SciCodetie28.828.5tie
AgentsEven
- OSWorld-VerifiedQwen3 VL 32B Reasoning by 10.430.641+10.4
- GDPValQwen3 VL 30B A3B Reasoning by 18.37.3+1
ReasoningQwen3 VL 32B Reasoning
- GPQA DiamondQwen3 VL 32B Reasoning by 1.37273.3+1.3
- Humanity's Last ExamQwen3 VL 32B Reasoning by 1.28.910.1+1.2
- CritPttie00tie
FactsQwen3 VL 32B Reasoning
- AA-Omniscience · Non-hallucinationQwen3 VL 32B Reasoning by 6.110.216.3+6.1
- AA-Omniscience · Accuracytie16.917tie
Images and chartsQwen3 VL 32B Reasoning
Long documentsQwen3 VL 32B Reasoning
- AA-LCRQwen3 VL 32B Reasoning by 12.642.755.3+12.6
Following instructionsQwen3 VL 32B Reasoning
- IFBenchQwen3 VL 32B Reasoning by 14.345.159.4+14.3
Other results32 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.
- SimpleQAQwen3 VL 32B Reasoning by 31.523.955.4+31.5
- τ²-Bench Telecom (AA run)Qwen3 VL 32B Reasoning by 25.719.945.6+25.7
- AA Agentic IndexQwen3 VL 32B Reasoning by 8.514.923.4+8.5
- ERQAQwen3 VL 32B Reasoning by 745.352.3+7
- IFEvalQwen3 VL 32B Reasoning by 6.181.787.8+6.1
- OCRBench v2_enQwen3 VL 32B Reasoning by 5.862.668.4+5.8
- AA-OmniscienceQwen3 VL 32B Reasoning by 5.1-57.6-52.5+5.1
- MATH-VisionQwen3 VL 32B Reasoning by 4.565.770.2+4.5
- VideoMMMUQwen3 VL 32B Reasoning by 47579+4
- Arena-Hard-V2Qwen3 VL 32B Reasoning by 3.856.760.5+3.8
- LVBenchQwen3 VL 32B Reasoning by 3.459.262.6+3.4
- InfoVQA_testQwen3 VL 32B Reasoning by 3.28689.2+3.2
- BFCL v3Qwen3 VL 32B Reasoning by 3.168.671.7+3.1
- SuperGPQAQwen3 VL 32B Reasoning by 2.656.459+2.6
- AA IntelligenceQwen3 VL 32B Reasoning by 2.49.511.9+2.4
- MMMU (val) (Pass@1)Qwen3 VL 32B Reasoning by 2.17678.1+2.1
- AI2DQwen3 VL 32B Reasoning by 286.988.9+2
- IncludeQwen3 VL 32B Reasoning by 1.874.576.3+1.8
- OCRBench v2 (Chinese)Qwen3 VL 32B Reasoning by 1.760.462.1+1.7
- MMLU-ProQwen3 VL 32B Reasoning by 1.680.582.1+1.6
- Artificial Analysis Coding IndexQwen3 VL 32B Reasoning by 1.413.114.5+1.4
- HallusionBenchQwen3 VL 32B Reasoning by 1.46667.4+1.4
- MV-BenchQwen3 VL 32B Reasoning by 1.27273.2+1.2
- DocVQA_testQwen3 VL 32B Reasoning by 1.19596.1+1.1
- MMLU-ProXQwen3 VL 32B Reasoning by 1.176.177.2+1.1
- MMLU-ReduxQwen3 VL 32B Reasoning by 190.991.9+1
- RealWorldQAQwen3 VL 32B Reasoning by 177.478.4+1
- ScreenSpotQwen3 VL 32B Reasoning by 194.795.7+1
- AIME 2025tie83.183.7tie
- MM MTBenchtie7.98.3tie
- ScreenSpot-Pro (No tools)tie57.357.1tie
- CharadesSTAtie62.762.8tie
Questions people ask
Which is better, Qwen3 VL 30B A3B Reasoning or Qwen3 VL 32B Reasoning?
Qwen3 VL 32B Reasoning wins six of the seven areas where both have results: coding, reasoning, facts, images and charts, long documents and following instructions. Qwen3 VL 30B A3B Reasoning wins none. They are level on agents.
Which is better for coding?
Qwen3 VL 32B Reasoning. It wins 2 of the 3 coding tests both models report; Qwen3 VL 30B A3B Reasoning wins none, and 1 is a tie.
Which is cheaper?
Qwen3 VL 30B A3B Reasoning costs $0.20 per million input tokens and $2.40 per million output tokens; Qwen3 VL 32B Reasoning costs $0.16 and $0.64. That makes Qwen3 VL 32B Reasoning about 69% cheaper for the same work.
How do you compare the two?
We use the 50 benchmark tests both models have published scores on. The verdict counts the 18 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 32 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.