Qwen3.6 35B A3B vs Qwen3.6 Plus
Wins 0 of 8 areas
—
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Qwen3.6 Plus is the stronger all-rounder.Qwen3.6 35B A3B is cheaper.
Scores updated · 52 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. 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 software07Qwen3.6 Plus7 of 7 tests
- MathCompetition and research-level math04Qwen3.6 Plus4 of 4 tests
- AgentsCarrying out multi-step tasks on its own03Qwen3.6 Plus3 of 3 tests
- ReasoningHard problems that need careful thinking03Qwen3.6 Plus3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video03Qwen3.6 Plus3 of 3 tests
- FactsGetting facts right instead of making them up02Qwen3.6 Plus2 of 2 tests
- Long documentsFinding answers in very long texts01Qwen3.6 Plus1 of 1 test
- Following instructionsDoing exactly what it is asked01Qwen3.6 Plus1 of 1 test
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.
Qwen3.6 35B A3B costs 25% 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 Qwen3.6 Plus pulls ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+17.9points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+16.5points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+15.9points ahead
Where Qwen3.6 35B A3B pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 52 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.6 Plus
- Terminal-Bench 2.1Qwen3.6 Plus by 16.544.961.4+16.5
- Terminal-Bench HardQwen3.6 Plus by 9.134.843.9+9.1
- SWE-bench ProQwen3.6 Plus by 7.149.556.6+7.1
- LiveCodeBench v6Qwen3.6 Plus by 6.780.487.1+6.7
- SWE-bench MultilingualQwen3.6 Plus by 6.667.273.8+6.6
- SWE-bench VerifiedQwen3.6 Plus by 5.473.478.8+5.4
- SciCodeQwen3.6 Plus by 4.136.640.7+4.1
AgentsQwen3.6 Plus
- τ-Bench V3 · BankingQwen3.6 Plus by 11.59.320.8+11.5
- MCP AtlasQwen3.6 Plus by 11.362.874.1+11.3
- GDPValQwen3.6 Plus by 4.919.824.7+4.9
ReasoningQwen3.6 Plus
- Humanity's Last ExamQwen3.6 Plus by 5.622.227.8+5.6
- GPQA DiamondQwen3.6 Plus by 4.184.188.2+4.1
- CritPtQwen3.6 Plus by 2.60.32.9+2.6
FactsQwen3.6 Plus
- AA-Omniscience · Non-hallucinationQwen3.6 Plus by 15.949.565.4+15.9
- AA-Omniscience · AccuracyQwen3.6 Plus by 7.618.826.4+7.6
Images and chartsQwen3.6 Plus
- MMMUQwen3.6 Plus by 4.381.786+4.3
- CharXiv (RQ)Qwen3.6 Plus by 3.57881.5+3.5
- MMMU-ProQwen3.6 Plus by 37578+3
MathQwen3.6 Plus
- FrontierMath Tiers 1-3 (v2)Qwen3.6 Plus by 17.920.438.3+17.9
- IMOAnswerBenchQwen3.6 Plus by 4.978.983.8+4.9
- HMMT Feb. 2026Qwen3.6 Plus by 4.283.687.8+4.2
- AIME 2026Qwen3.6 Plus by 2.692.795.3+2.6
Following instructionsQwen3.6 Plus
- IFBenchQwen3.6 Plus by 10.864.475.2+10.8
Other results28 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.
- AA-OmniscienceQwen3.6 Plus by 23.1-22.20.9+23.1
- DeepPlanningQwen3.6 Plus by 15.625.941.5+15.6
- WideSearchQwen3.6 Plus by 14.260.174.3+14.2
- AA Agentic IndexQwen3.6 Plus by 141529+14
- Tool-DecathlonQwen3.6 Plus by 12.926.939.8+12.9
- ToolathlonQwen3.6 Plus by 12.926.939.8+12.9
- Artificial Analysis Coding IndexQwen3.6 Plus by 12.641.954.5+12.6
- MCPMarkQwen3.6 Plus by 11.23748.2+11.2
- Terminal-Bench 2.0Qwen3.6 Plus by 10.151.561.6+10.1
- AA IntelligenceQwen3.6 Plus by 8.818.227+8.8
- Claw Eval (pass@3)Qwen3.6 Plus by 8.75058.7+8.7
- VITA-BenchQwen3.6 Plus by 8.735.644.3+8.7
- NL2RepoQwen3.6 Plus by 8.529.437.9+8.5
- SimpleVQAQwen3.6 Plus by 8.458.967.3+8.4
- SuperGPQAQwen3.6 Plus by 6.964.771.6+6.9
- HMMT 2025Qwen3.6 Plus by 5.589.194.6+5.5
- HMMT Nov. 2025Qwen3.6 Plus by 5.589.194.6+5.5
- τ³-BenchQwen3.6 Plus by 3.567.270.7+3.5
- C-EvalQwen3.6 Plus by 3.39093.3+3.3
- MMLU-ProQwen3.6 Plus by 3.385.288.5+3.3
- τ²-Bench Telecom (AA run)Qwen3.6 Plus by 2.495.397.7+2.4
- AI2DQwen3.6 Plus by 1.792.794.4+1.7
- CC-OCRQwen3.6 Plus by 1.581.983.4+1.5
- OmniDocBench 1.5Qwen3.6 Plus by 1.389.991.2+1.3
- MMLU-ReduxQwen3.6 Plus by 1.293.394.5+1.2
- MLVUtie86.286.7tie
- VideoMMMUtie83.784tie
- RealWorldQAtie85.385.4tie
Questions people ask
Which is better, Qwen3.6 35B A3B or Qwen3.6 Plus?
Qwen3.6 Plus wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. Qwen3.6 35B A3B wins none, but costs 25% less.
Which is better for coding?
Qwen3.6 Plus. It wins 7 of the 7 coding tests both models report; Qwen3.6 35B A3B wins none.
Which is cheaper?
Qwen3.6 35B A3B costs $0.38 per million input tokens and $2.25 per million output tokens; Qwen3.6 Plus costs $0.50 and $3.00. That makes Qwen3.6 35B A3B about 25% cheaper for the same work.
How do you compare the two?
We use the 52 benchmark tests both models have published scores on. The verdict counts the 24 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 28 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.