Qwen3.6 Plus vs Qwen3.7 Max
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Coding · Agents · Reasoning · Facts · Math · Following instructions
Qwen3.7 Max is the stronger all-rounder.Qwen3.6 Plus is cheaper.
Scores updated · 52 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 software19Qwen3.7 Max9 of 10 tests
- MathCompetition and research-level math05Qwen3.7 Max5 of 5 tests
- ReasoningHard problems that need careful thinking04Qwen3.7 Max4 of 4 tests
- FactsGetting facts right instead of making them up03Qwen3.7 Max3 of 3 tests
- Following instructionsDoing exactly what it is asked02Qwen3.7 Max2 of 2 tests
- AgentsCarrying out multi-step tasks on its own12Qwen3.7 Max2 of 3 tests
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
Long documents rests on a single test.
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 Plus costs 65% 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 Qwen3.7 Max pulls ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+26.3points ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+15.7points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+13.1points ahead
Where Qwen3.6 Plus pulls ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+9points ahead
- Programming problems, refreshed regularlyLiveBench · Coding+4points ahead
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.
CodingQwen3.7 Max
- Terminal-Bench 2.1Qwen3.7 Max by 13.161.474.5+13.1
- SciCodeQwen3.7 Max by 8.840.749.5+8.8
- Terminal-Bench HardQwen3.7 Max by 6.943.950.8+6.9
- LMArena · WebDevQwen3.7 Max by 54 rating points14611515+54 rating
- LiveCodeBench v6Qwen3.7 Max by 4.587.191.6+4.5
- SWE-bench MultilingualQwen3.7 Max by 4.573.878.3+4.5
- LiveBench · CodingQwen3.6 Plus by 478.274.2+4
- SWE-bench ProQwen3.7 Max by 456.660.6+4
- LiveBench · Agentic CodingQwen3.7 Max by 2.241.443.6+2.2
- SWE-bench VerifiedQwen3.7 Max by 1.678.880.4+1.6
AgentsQwen3.7 Max
- τ-Bench V3 · BankingQwen3.6 Plus by 920.811.8+9
- GDPValQwen3.7 Max by 6.924.731.6+6.9
- MCP AtlasQwen3.7 Max by 2.374.176.4+2.3
ReasoningQwen3.7 Max
- Humanity's Last ExamQwen3.7 Max by 12.727.840.5+12.7
- CritPtQwen3.7 Max by 10.52.913.4+10.5
- LiveBench · ReasoningQwen3.7 Max by 7.575.883.3+7.5
- GPQA DiamondQwen3.7 Max by 4.188.292.3+4.1
FactsQwen3.7 Max
- SimpleQA VerifiedQwen3.7 Max by 11.744.155.8+11.7
- AA-Omniscience · Non-hallucinationQwen3.7 Max by 965.474.4+9
- AA-Omniscience · AccuracyQwen3.7 Max by 4.726.431.1+4.7
MathQwen3.7 Max
- FrontierMath Tiers 1-3 (v2)Qwen3.7 Max by 26.338.364.6+26.3
- HMMT Feb. 2026Qwen3.7 Max by 9.387.897.1+9.3
- IMOAnswerBenchQwen3.7 Max by 6.283.890+6.2
- AIME 2026Qwen3.7 Max by 1.795.397+1.7
- LiveBench · MathematicsQwen3.7 Max by 1.683.785.3+1.6
Following instructionsQwen3.7 Max
- LiveBench · Instruction FollowingQwen3.7 Max by 15.758.374+15.7
- IFBenchQwen3.7 Max by 5.375.280.5+5.3
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.
- AA-OmniscienceQwen3.7 Max by 12.60.913.5+12.6
- MCPMarkQwen3.7 Max by 12.648.260.8+12.6
- Artificial Analysis Coding IndexQwen3.7 Max by 11.554.566+11.5
- NL2RepoQwen3.7 Max by 9.337.947.2+9.3
- Terminal-Bench 2.0Qwen3.7 Max by 8.161.669.7+8.1
- Finance Agent v2Qwen3.7 Max by 7.540.948.4+7.5
- Claw Eval (pass@3)Qwen3.7 Max by 6.558.765.2+6.5
- AA Agentic IndexQwen3.6 Plus by 5.12923.9+5.1
- livebench_languageQwen3.7 Max by 4.77579.7+4.7
- VITA-BenchQwen3.7 Max by 3.644.347.9+3.6
- LiveBenchQwen3.7 Max by 3.470.874.3+3.4
- τ²-Bench Telecom (AA run)Qwen3.6 Plus by 397.794.7+3
- HLE (with tools)Qwen3.7 Max by 2.950.653.5+2.9
- AA IntelligenceQwen3.7 Max by 2.52729.5+2.5
- MMLU-ProXQwen3.7 Max by 2.384.787+2.3
- SuperGPQAQwen3.7 Max by 271.673.6+2
- livebench_data_analysisQwen3.7 Max by 1.969.971.8+1.9
- IncludeQwen3.7 Max by 1.185.186.2+1.1
- MMLU-ProQwen3.7 Max by 1.188.589.6+1.1
- WideSearchtie74.375.2tie
- MMMLUtie89.590.3tie
- MMLU-Reduxtie94.595tie
- HMMT Nov. 2025tie94.695tie
- IFEvaltie94.394.3tie
Questions people ask
Which is better, Qwen3.6 Plus or Qwen3.7 Max?
Qwen3.7 Max wins six of the seven areas where both have results: coding, agents, reasoning, facts, math and following instructions. Qwen3.6 Plus wins none, but costs 65% less. They are level on long documents.
Which is better for coding?
Qwen3.7 Max. It wins 9 of the 10 coding tests both models report; Qwen3.6 Plus wins 1.
Which is cheaper?
Qwen3.6 Plus costs $0.50 per million input tokens and $3.00 per million output tokens; Qwen3.7 Max costs $2.50 and $7.50. That makes Qwen3.6 Plus about 65% 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 28 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.