Claude Opus 4.5 vs Qwen3.6 27B
Wins 4 of 8 areas
Coding · Agents · Reasoning · Math
Wins 1 of 8 areas
Images and charts
Claude Opus 4.5 wins more areas, narrowly.Qwen3.6 27B is cheaper and better at images and charts.
Scores updated · 49 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.
- MathCompetition and research-level math40Claude Opus 4.54 of 5 tests · 1 tie
- ReasoningHard problems that need careful thinking40Claude Opus 4.54 of 4 tests
- CodingWriting and fixing software42Claude Opus 4.54 of 8 tests · 2 ties
- AgentsCarrying out multi-step tasks on its own10Claude Opus 4.51 of 1 test
- Images and chartsUnderstanding pictures, charts and video04Qwen3.6 27B4 of 5 tests · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
- Following instructionsDoing exactly what it is asked11Even1 each
Agents and long documents 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 27B costs 86% 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
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+27points ahead
- Real work tasks from 44 professionsGDPVal+22.9points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+12.2points ahead
Where Qwen3.6 27B pulls ahead
- Fixes real GitHub issues in many programming languagesSWE-bench Multilingual+64.3points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+11.7points ahead
- Reasoning about charts from research papersCharXiv (RQ)+11.2points ahead
Every test, side by side
All 49 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4.5
- SWE-bench MultilingualQwen3.6 27B by 64.3771.3+64.3
- Terminal-Bench HardClaude Opus 4.5 by 12.24734.8+12.2
- LiveBench · CodingClaude Opus 4.5 by 7.979.771.8+7.9
- SciCodeClaude Opus 4.5 by 6.749.542.8+6.7
- SWE-bench VerifiedClaude Opus 4.5 by 3.780.977.2+3.7
- SWE-bench ProQwen3.6 27B by 1.55253.5+1.5
- LiveCodeBench v6tie84.883.9tie
- LiveBench · Agentic Codingtie39.739.3tie
ReasoningClaude Opus 4.5
- LiveBench · ReasoningClaude Opus 4.5 by 9.880.170.3+9.8
- Humanity's Last ExamClaude Opus 4.5 by 730.123.1+7
- CritPtClaude Opus 4.5 by 3.54.61.1+3.5
- GPQA DiamondClaude Opus 4.5 by 2.486.684.2+2.4
FactsEven
- AA-Omniscience · AccuracyClaude Opus 4.5 by 2746.619.6+27
- AA-Omniscience · Non-hallucinationQwen3.6 27B by 11.73950.7+11.7
Images and chartsQwen3.6 27B
- CharXiv (RQ)Qwen3.6 27B by 11.267.278.4+11.2
- MathVistaQwen3.6 27B by 7.280.287.4+7.2
- OCRBenchQwen3.6 27B by 2.986.589.4+2.9
- MMMUQwen3.6 27B by 2.280.782.9+2.2
- MMMU-Protie7474.6tie
MathClaude Opus 4.5
- LiveBench · MathematicsClaude Opus 4.5 by 10.590.479.9+10.5
- IMOAnswerBenchClaude Opus 4.5 by 3.28480.8+3.2
- AIME 2026Claude Opus 4.5 by 195.194.1+1
- HMMT Feb. 2026Claude Opus 4.5 by 185.384.3+1
- FrontierMath Tiers 1-3 (v2)tie34.435.1tie
Following instructionsEven
- IFBenchQwen3.6 27B by 9.65867.6+9.6
- LiveBench · Instruction FollowingClaude Opus 4.5 by 9.462.553.2+9.4
Other results21 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 39.559.620.1+39.5
- AA-OmniscienceClaude Opus 4.5 by 3414-20+34
- livebench_languageClaude Opus 4.5 by 1881.363.3+18
- SimpleVQAClaude Opus 4.5 by 13.669.756.1+13.6
- AA IntelligenceClaude Opus 4.5 by 7.729.121.4+7.7
- NL2RepoClaude Opus 4.5 by 743.236.2+7
- Artificial Analysis Coding IndexQwen3.6 27B by 5.947.853.7+5.9
- QwenWebBenchClaude Opus 4.5 by 49 rating points15361487+49 rating
- τ²-Bench Telecom (AA run)Qwen3.6 27B by 4.789.594.2+4.7
- SuperGPQAClaude Opus 4.5 by 4.670.666+4.6
- Claw-Eval AvgClaude Opus 4.5 by 4.276.672.4+4.2
- livebench_data_analysisClaude Opus 4.5 by 474.470.4+4
- MMLU-ProClaude Opus 4.5 by 3.389.586.2+3.3
- SkillsBench Avg5Qwen3.6 27B by 2.945.348.2+2.9
- HMMT Nov. 2025Claude Opus 4.5 by 2.693.390.7+2.6
- MMLU-ReduxClaude Opus 4.5 by 2.195.693.5+2.1
- QwenClawBenchQwen3.6 27B by 1.152.353.4+1.1
- HMMT Feb. 2025tie92.993.8tie
- C-Evaltie92.291.4tie
- Terminal-Bench 2.0tie59.359.3tie
- VideoMMMUtie84.484.4tie
Questions people ask
Which is better, Claude Opus 4.5 or Qwen3.6 27B?
Claude Opus 4.5 wins four of the eight areas we test: coding, agents, reasoning and math. Qwen3.6 27B wins images and charts, and costs 86% less. They are level on facts, long documents and following instructions.
Which is better for coding?
Claude Opus 4.5. It wins 4 of the 8 coding tests both models report; Qwen3.6 27B wins 2, and 2 are ties.
Which is cheaper?
Claude Opus 4.5 costs $5.00 per million input tokens and $25.00 per million output tokens; Qwen3.6 27B costs $0.60 and $3.60. That makes Qwen3.6 27B about 86% cheaper for the same work.
How do you compare the two?
We use the 49 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 21 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.