The two are evenly matched.Qwen3.5 2B is better at long documents; Qwen3 VL 4B Instruct at images and charts.
Scores updated · 33 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 video04Qwen3 VL 4B Instruct4 of 5 tests · 1 tie
- Long documentsFinding answers in very long texts10Qwen3.5 2B1 of 1 test
- CodingWriting and fixing software11Even1 each
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
- ReasoningHard problems that need careful thinking11Even1 each · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Agents, long documents and following instructions rest on a single test each.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Qwen3 VL 4B Instruct pulls ahead
Where Qwen3.5 2B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+23.2points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+8.5points ahead
- Reasons across sets of long documentsAA-LCR+7points ahead
Every test, side by side
All 33 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- SciCodeQwen3 VL 4B Instruct by 10.92.813.7+10.9
- Terminal-Bench HardQwen3.5 2B by 3.83.80+3.8
ReasoningEven
- GPQA DiamondQwen3.5 2B by 8.545.637.1+8.5
- Humanity's Last ExamQwen3 VL 4B Instruct by 12.63.6+1
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationQwen3.5 2B by 23.225.82.6+23.2
- AA-Omniscience · AccuracyQwen3 VL 4B Instruct by 2.88.110.9+2.8
Images and chartsQwen3 VL 4B Instruct
Other results18 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.
- τ²-Bench Telecom (AA run)Qwen3.5 2B by 45.66923.4+45.6
- AIME 2025Qwen3 VL 4B Instruct by 34.61246.6+34.6
- MMMU (val) (Pass@1)Qwen3 VL 4B Instruct by 23.344.167.4+23.3
- AA-OmniscienceQwen3.5 2B by 15.8-60.2-76+15.8
- OCRBench v2_enQwen3 VL 4B Instruct by 15.648.163.7+15.6
- HallusionBenchQwen3.5 2B by 7.965.557.6+7.9
- MMLU-ProXQwen3 VL 4B Instruct by 7.152.359.4+7.1
- IncludeQwen3 VL 4B Instruct by 655.461.4+6
- RealWorldQAQwen3 VL 4B Instruct by 5.865.170.9+5.8
- IFEvalQwen3 VL 4B Instruct by 3.778.682.3+3.7
- TextVQA-valQwen3 VL 4B Instruct by 3.377.380.5+3.3
- SuperGPQAQwen3 VL 4B Instruct by 2.837.540.3+2.8
- VLM Judge ScoreQwen3 VL 4B Instruct by 2.389.792+2.3
- DocVQA-valQwen3 VL 4B Instruct by 2.192.694.7+2.1
- MMLU-ReduxQwen3 VL 4B Instruct by 1.979.681.5+1.9
- Artificial Analysis Coding IndexQwen3 VL 4B Instruct by 1.72.94.6+1.7
- AA IntelligenceQwen3.5 2B by 1.26.95.7+1.2
- MMLU-Protie66.567.1tie
Questions people ask
Which is better, Qwen3.5 2B or Qwen3 VL 4B Instruct?
Qwen3.5 2B and Qwen3 VL 4B Instruct each win one of the seven areas where both have results. Qwen3.5 2B wins long documents; Qwen3 VL 4B Instruct wins images and charts. They are level on coding, agents, reasoning, facts and following instructions.
Which is better for coding?
Neither. They win 1 coding test each of the 2 both models report.
How do you compare the two?
We use the 33 benchmark tests both models have published scores on. The verdict counts the 15 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.