Claude Opus 4.5 vs Qwen3.6 Plus
Wins 3 of 8 areas
Agents · Reasoning · Facts
Wins 2 of 8 areas
Images and charts · Math
Claude Opus 4.5 wins more areas, narrowly.Qwen3.6 Plus is cheaper and better at images and charts.
Scores updated · 54 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.
- ReasoningHard problems that need careful thinking31Claude Opus 4.53 of 4 tests
- AgentsCarrying out multi-step tasks on its own21Claude Opus 4.52 of 3 tests
- FactsGetting facts right instead of making them up21Claude Opus 4.52 of 3 tests
- Images and chartsUnderstanding pictures, charts and video03Qwen3.6 Plus3 of 3 tests
- MathCompetition and research-level math12Qwen3.6 Plus2 of 5 tests · 2 ties
- CodingWriting and fixing software44Even4 each · 1 tie
- Long documentsFinding answers in very long texts11Even1 each
- Following instructionsDoing exactly what it is asked11Even1 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 Plus costs 88% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Claude Opus 4.5 pulls ahead
- Real work tasks from 44 professionsGDPVal+22.6points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+20.2points ahead
- Code for real scientific research problemsSciCode+8.8points ahead
Where Qwen3.6 Plus pulls ahead
- Fixes real GitHub issues in many programming languagesSWE-bench Multilingual+66.8points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+26.4points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+17.2points ahead
Every test, side by side
All 54 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- SWE-bench MultilingualQwen3.6 Plus by 66.8773.8+66.8
- SciCodeClaude Opus 4.5 by 8.849.540.7+8.8
- SWE-bench ProQwen3.6 Plus by 4.65256.6+4.6
- Terminal-Bench HardClaude Opus 4.5 by 3.14743.9+3.1
- LiveCodeBench v6Qwen3.6 Plus by 2.384.887.1+2.3
- SWE-bench VerifiedClaude Opus 4.5 by 2.180.978.8+2.1
- LiveBench · Agentic CodingQwen3.6 Plus by 1.739.741.4+1.7
- LiveBench · CodingClaude Opus 4.5 by 1.579.778.2+1.5
- LMArena · WebDevtie14681461tie
AgentsClaude Opus 4.5
- GDPValClaude Opus 4.5 by 22.647.324.7+22.6
- MCP AtlasQwen3.6 Plus by 11.862.374.1+11.8
- OSWorld-VerifiedClaude Opus 4.5 by 3.866.362.5+3.8
ReasoningClaude Opus 4.5
- LiveBench · ReasoningClaude Opus 4.5 by 4.380.175.8+4.3
- Humanity's Last ExamClaude Opus 4.5 by 2.330.127.8+2.3
- CritPtClaude Opus 4.5 by 1.74.62.9+1.7
- GPQA DiamondQwen3.6 Plus by 1.686.688.2+1.6
FactsClaude Opus 4.5
- AA-Omniscience · Non-hallucinationQwen3.6 Plus by 26.43965.4+26.4
- AA-Omniscience · AccuracyClaude Opus 4.5 by 20.246.626.4+20.2
- SimpleQA VerifiedClaude Opus 4.5 by 1.645.744.1+1.6
Images and chartsQwen3.6 Plus
- CharXiv (RQ)Qwen3.6 Plus by 14.367.281.5+14.3
- MMMUQwen3.6 Plus by 5.380.786+5.3
- MMMU-ProQwen3.6 Plus by 47478+4
MathQwen3.6 Plus
- LiveBench · MathematicsClaude Opus 4.5 by 6.790.483.7+6.7
- FrontierMath Tiers 1-3 (v2)Qwen3.6 Plus by 3.934.438.3+3.9
- HMMT Feb. 2026Qwen3.6 Plus by 2.585.387.8+2.5
- AIME 2026tie95.195.3tie
- IMOAnswerBenchtie8483.8tie
Long documentsEven
- LongBench v2Claude Opus 4.5 by 2.464.462+2.4
- AA-LCRQwen3.6 Plus by 177.378.3+1
Following instructionsEven
- IFBenchQwen3.6 Plus by 17.25875.2+17.2
- LiveBench · Instruction FollowingClaude Opus 4.5 by 4.362.558.3+4.3
Other results23 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 Agentic IndexClaude Opus 4.5 by 30.659.629+30.6
- AA-OmniscienceClaude Opus 4.5 by 13.1140.9+13.1
- MATH-VisionQwen3.6 Plus by 10.977.188+10.9
- τ²-Bench Telecom (AA run)Qwen3.6 Plus by 8.289.597.7+8.2
- HLE (with tools)Qwen3.6 Plus by 7.443.250.6+7.4
- Artificial Analysis Coding IndexQwen3.6 Plus by 6.747.854.5+6.7
- livebench_languageClaude Opus 4.5 by 6.381.375+6.3
- NL2RepoClaude Opus 4.5 by 5.343.237.9+5.3
- LiveBenchClaude Opus 4.5 by 5.17670.8+5.1
- livebench_data_analysisClaude Opus 4.5 by 4.574.469.9+4.5
- frontiermath_tier_4_v1Qwen3.6 Plus by 4.14.28.3+4.1
- ToolathlonClaude Opus 4.5 by 3.743.539.8+3.7
- OmniDocBench 1.5Qwen3.6 Plus by 3.587.791.2+3.5
- SimpleVQAClaude Opus 4.5 by 2.469.767.3+2.4
- Terminal-Bench 2.0Qwen3.6 Plus by 2.359.361.6+2.3
- AA IntelligenceClaude Opus 4.5 by 2.129.127+2.1
- HMMT Nov. 2025Qwen3.6 Plus by 1.393.394.6+1.3
- MMMLUClaude Opus 4.5 by 1.390.889.5+1.3
- C-EvalQwen3.6 Plus by 1.192.293.3+1.1
- MMLU-ReduxClaude Opus 4.5 by 1.195.694.5+1.1
- MMLU-ProClaude Opus 4.5 by 189.588.5+1
- SuperGPQAQwen3.6 Plus by 170.671.6+1
- VideoMMMUtie84.484tie
Questions people ask
Which is better, Claude Opus 4.5 or Qwen3.6 Plus?
Claude Opus 4.5 wins three of the eight areas we test: agents, reasoning and facts. Qwen3.6 Plus wins images and charts and math, and costs 88% less. They are level on coding, long documents and following instructions.
Which is better for coding?
Neither. They win 4 coding tests each of the 9 both models report, and 1 is a tie.
Which is cheaper?
Claude Opus 4.5 costs $5.00 per million input tokens and $25.00 per million output tokens; Qwen3.6 Plus costs $0.50 and $3.00. That makes Qwen3.6 Plus about 88% cheaper for the same work.
How do you compare the two?
We use the 54 benchmark tests both models have published scores on. The verdict counts the 31 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 23 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.