Qwen3.5 0.8B vs Qwen3.5 2B
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
Qwen3.5 2B is the stronger all-rounder.
Scores updated · 39 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.
- CodingWriting and fixing software03Qwen3.5 2B3 of 3 tests
- ReasoningHard problems that need careful thinking02Qwen3.5 2B2 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts02Qwen3.5 2B2 of 2 tests
- Following instructionsDoing exactly what it is asked02Qwen3.5 2B2 of 2 tests
- AgentsCarrying out multi-step tasks on its own01Qwen3.5 2B1 of 2 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video01Qwen3.5 2B1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Images and charts 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 Qwen3.5 2B pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+34.5points ahead
- Harder college exam questions with imagesMMMU-Pro+17.2points ahead
- Keeps track of context across a multi-turn chatMulti-Challenge+14.8points ahead
Where Qwen3.5 0.8B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+12.9points ahead
Every test, side by side
All 39 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.5 2B
- Terminal-Bench HardQwen3.5 2B by 3.803.8+3.8
- Terminal-Bench 2.1Qwen3.5 2B by 303+3
- SciCodeQwen3.5 2B by 2.802.8+2.8
ReasoningQwen3.5 2B
- GPQA DiamondQwen3.5 2B by 34.511.145.6+34.5
- Humanity's Last ExamQwen3.5 2B by 1.51.12.6+1.5
- CritPttie00tie
FactsEven
- AA-Omniscience · Non-hallucinationQwen3.5 0.8B by 12.938.725.8+12.9
- AA-Omniscience · AccuracyQwen3.5 2B by 3.94.28.1+3.9
Long documentsQwen3.5 2B
- LongBench v2Qwen3.5 2B by 12.626.138.7+12.6
- AA-LCRQwen3.5 2B by 12921+12
Following instructionsQwen3.5 2B
- Multi-ChallengeQwen3.5 2B by 14.818.933.7+14.8
- IFBenchQwen3.5 2B by 1021.531.5+10
Other results24 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.
- t2-benchQwen3.5 2B by 37.211.648.8+37.2
- MMLU-ProQwen3.5 2B by 24.242.366.5+24.2
- MATH-500 (EM)Qwen3.5 2B by 23.957.681.5+23.9
- C-EvalQwen3.5 2B by 22.750.573.2+22.7
- τ²-Bench Telecom (AA run)Qwen3.5 2B by 21.347.769+21.3
- MMLU-ReduxQwen3.5 2B by 20.159.579.6+20.1
- HumanEval+Qwen3.5 2B by 19.246.866+19.2
- MMMLUQwen3.5 2B by 18.844.363.1+18.8
- MMLU-ProXQwen3.5 2B by 17.734.652.3+17.7
- MMLU-ProX-LiteQwen3.5 2B by 16.338.955.2+16.3
- SuperGPQAQwen3.5 2B by 16.221.337.5+16.2
- MBPP+Qwen3.5 2B by 15.342.357.6+15.3
- IncludeQwen3.5 2B by 14.840.655.4+14.8
- MGSM-rev2Qwen3.5 2B by 13.672.986.5+13.6
- AIME 2024Qwen3.5 2B by 11.35.717+11.3
- Global-MMLU-LiteQwen3.5 2B by 11.253.364.5+11.2
- AIME 2025Qwen3.5 2B by 10.38.719+10.3
- BFCLv4Qwen3.5 2B by 8.625.333.9+8.6
- VLM Judge ScoreQwen3.5 2B by 7.482.389.7+7.4
- AA-OmniscienceQwen3.5 0.8B by 5.7-54.5-60.2+5.7
- BFCL v2Qwen3.5 2B by 4.96468.9+4.9
- Artificial Analysis Coding IndexQwen3.5 2B by 2.902.9+2.9
- IFEvalQwen3.5 2B by 2.471.273.6+2.4
- AA Intelligencetie6.16.9tie
Questions people ask
Which is better, Qwen3.5 0.8B or Qwen3.5 2B?
Qwen3.5 2B wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Qwen3.5 0.8B wins none. They are level on facts.
Which is better for coding?
Qwen3.5 2B. It wins 3 of the 3 coding tests both models report; Qwen3.5 0.8B wins none.
How do you compare the two?
We use the 39 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 24 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.