Qwen3.6 Plus vs Qwen3.7 Plus Preview
Wins 2 of 8 areas
Agents · Long documents
Wins 5 of 8 areas
Coding · Reasoning · Images and charts · Math · Following instructions
Qwen3.7 Plus Preview is the stronger all-rounder.Qwen3.6 Plus is better at agents.
Scores updated · 51 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 software15Qwen3.7 Plus Preview5 of 7 tests · 1 tie
- ReasoningHard problems that need careful thinking03Qwen3.7 Plus Preview3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video03Qwen3.7 Plus Preview3 of 3 tests
- MathCompetition and research-level math12Qwen3.7 Plus Preview2 of 3 tests
- Following instructionsDoing exactly what it is asked01Qwen3.7 Plus Preview1 of 1 test
- AgentsCarrying out multi-step tasks on its own21Qwen3.6 Plus2 of 4 tests · 1 tie
- Long documentsFinding answers in very long texts10Qwen3.6 Plus1 of 1 test
- FactsGetting facts right instead of making them up11Even1 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.
Qwen3.7 Plus Preview costs 43% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Qwen3.7 Plus Preview pulls ahead
- Completes tasks by operating a computer desktopOSWorld-Verified+10.8points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+7.8points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+6.9points ahead
Where Qwen3.6 Plus pulls ahead
- Real work tasks from 44 professionsGDPVal+11.2points ahead
- Reasons across sets of long documentsAA-LCR+5.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+3.9points ahead
Every test, side by side
All 51 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.7 Plus Preview
- SciCodeQwen3.7 Plus Preview by 5.440.746.1+5.4
- Terminal-Bench HardQwen3.7 Plus Preview by 3.143.947+3.1
- LiveCodeBench v6Qwen3.7 Plus Preview by 2.587.189.6+2.5
- SWE-bench MultilingualQwen3.7 Plus Preview by 273.875.8+2
- SWE-bench VerifiedQwen3.6 Plus by 1.178.877.7+1.1
- SWE-bench ProQwen3.7 Plus Preview by 156.657.6+1
- Terminal-Bench 2.1tie61.461tie
AgentsQwen3.6 Plus
- GDPValQwen3.6 Plus by 11.224.713.5+11.2
- OSWorld-VerifiedQwen3.7 Plus Preview by 10.862.573.3+10.8
- τ-Bench V3 · BankingQwen3.6 Plus by 3.320.817.5+3.3
- MCP Atlastie74.173.2tie
ReasoningQwen3.7 Plus Preview
- Humanity's Last ExamQwen3.7 Plus Preview by 7.827.835.6+7.8
- CritPtQwen3.7 Plus Preview by 6.22.99.1+6.2
- GPQA DiamondQwen3.7 Plus Preview by 1.888.290+1.8
FactsEven
- AA-Omniscience · Non-hallucinationQwen3.7 Plus Preview by 6.965.472.3+6.9
- AA-Omniscience · AccuracyQwen3.6 Plus by 3.926.422.5+3.9
Images and chartsQwen3.7 Plus Preview
- CharXiv (RQ)Qwen3.7 Plus Preview by 4.481.585.9+4.4
- Video-MMEQwen3.7 Plus Preview by 3.884.288+3.8
- MMMU-ProQwen3.7 Plus Preview by 2.57880.5+2.5
MathQwen3.7 Plus Preview
- HMMT Feb. 2026Qwen3.7 Plus Preview by 5.187.892.9+5.1
- FrontierMath Tiers 1-3 (v2)Qwen3.6 Plus by 3.938.334.4+3.9
- IMOAnswerBenchQwen3.7 Plus Preview by 2.283.886+2.2
Following instructionsQwen3.7 Plus Preview
- IFBenchQwen3.7 Plus Preview by 2.875.278+2.8
Other results27 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.
- DeepPlanningQwen3.7 Plus Preview by 20.841.562.3+20.8
- SimpleVQAQwen3.7 Plus Preview by 14.467.381.7+14.4
- ScreenSpot-Pro (No tools)Qwen3.7 Plus Preview by 10.868.279+10.8
- MCPMarkQwen3.7 Plus Preview by 10.548.258.7+10.5
- AA Agentic IndexQwen3.6 Plus by 9.32919.7+9.3
- Terminal-Bench 2.0Qwen3.7 Plus Preview by 8.761.670.3+8.7
- τ²-Bench Telecom (AA run)Qwen3.6 Plus by 4.797.793+4.7
- ERQAQwen3.7 Plus Preview by 4.165.769.8+4.1
- Claw Eval (pass@3)Qwen3.7 Plus Preview by 458.762.7+4
- NL2RepoQwen3.7 Plus Preview by 3.237.941.1+3.2
- Finance Agent v2Qwen3.6 Plus by 2.740.938.2+2.7
- MATH-VisionQwen3.7 Plus Preview by 2.38890.3+2.3
- IncludeQwen3.6 Plus by 2.185.183+2.1
- AA IntelligenceQwen3.6 Plus by 1.82725.2+1.8
- RealWorldQAQwen3.7 Plus Preview by 1.585.486.9+1.5
- Artificial Analysis Coding IndexQwen3.7 Plus Preview by 1.454.555.9+1.4
- VideoMMMUQwen3.7 Plus Preview by 1.48485.4+1.4
- VITA-BenchQwen3.7 Plus Preview by 1.344.345.6+1.3
- MLVUtie86.787.4tie
- MMLU-ProXtie84.785.4tie
- MMMLUtie89.589tie
- IFEvaltie94.394.6tie
- AA-Omnisciencetie0.91.1tie
- OmniDocBench 1.5tie91.291.4tie
- SuperGPQAtie71.671.4tie
- MMLU-Protie88.588.5tie
- MMLU-Reduxtie94.594.5tie
Questions people ask
Which is better, Qwen3.6 Plus or Qwen3.7 Plus Preview?
Qwen3.7 Plus Preview wins five of the eight areas we test: coding, reasoning, images and charts, math and following instructions. Qwen3.6 Plus wins agents and long documents. They are level on facts.
Which is better for coding?
Qwen3.7 Plus Preview. It wins 5 of the 7 coding tests both models report; Qwen3.6 Plus wins 1, and 1 is a tie.
Which is cheaper?
Qwen3.6 Plus costs $0.50 per million input tokens and $3.00 per million output tokens; Qwen3.7 Plus Preview costs $0.40 and $1.60. That makes Qwen3.7 Plus Preview about 43% cheaper for the same work.
How do you compare the two?
We use the 51 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 27 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.