Claude Opus 4.6 vs Qwen3.5 27B
Wins 5 of 8 areas
Coding · Agents · Reasoning · Facts · Math
Wins 1 of 8 areas
Following instructions
Claude Opus 4.6 is the stronger all-rounder.Qwen3.5 27B is cheaper and better at following instructions.
Scores updated · 48 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 software70Claude Opus 4.67 of 7 tests
- ReasoningHard problems that need careful thinking30Claude Opus 4.63 of 3 tests
- AgentsCarrying out multi-step tasks on its own31Claude Opus 4.63 of 4 tests
- FactsGetting facts right instead of making them up20Claude Opus 4.62 of 3 tests · 1 tie
- MathCompetition and research-level math21Claude Opus 4.62 of 3 tests
- Following instructionsDoing exactly what it is asked02Qwen3.5 27B2 of 2 tests
- Images and chartsUnderstanding pictures, charts and video11Even1 each · 1 tie
- 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.
Qwen3.5 27B costs 91% 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 4.6 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+26.3points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+23points ahead
- Real work tasks from 44 professionsGDPVal+22.9points ahead
Where Qwen3.5 27B pulls ahead
- Keeps track of context across a multi-turn chatMulti-Challenge+23.6points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+22.5points ahead
- Multi-step tasks using many real software toolsMCP Atlas+5.7points ahead
Every test, side by side
All 48 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4.6
- LMArena · WebDevClaude Opus 4.6 by 189 rating points15461357+189 rating
- Terminal-Bench HardClaude Opus 4.6 by 13.646.232.6+13.6
- SciCodeClaude Opus 4.6 by 12.451.939.5+12.4
- SWE-bench MultilingualClaude Opus 4.6 by 8.577.869.3+8.5
- SWE-bench VerifiedClaude Opus 4.6 by 8.480.872.4+8.4
- LiveCodeBench v6Claude Opus 4.6 by 8.188.880.7+8.1
- SWE-bench ProClaude Opus 4.6 by 2.253.451.2+2.2
AgentsClaude Opus 4.6
- BrowseCompClaude Opus 4.6 by 238461+23
- GDPValClaude Opus 4.6 by 22.955.933+22.9
- OSWorld-VerifiedClaude Opus 4.6 by 16.572.756.2+16.5
- MCP AtlasQwen3.5 27B by 5.762.768.4+5.7
ReasoningClaude Opus 4.6
- Humanity's Last ExamClaude Opus 4.6 by 1639.923.9+16
- CritPtClaude Opus 4.6 by 11.712.60.9+11.7
- GPQA DiamondClaude Opus 4.6 by 3.889.685.8+3.8
FactsClaude Opus 4.6
- AA-Omniscience · AccuracyClaude Opus 4.6 by 26.34720.7+26.3
- AA-Omniscience · Non-hallucinationClaude Opus 4.6 by 18.737.218.5+18.7
- Vectara HHEM hallucination ratelower is bettertie12.212.1tie
Images and chartsEven
- LMArena · VisionClaude Opus 4.6 by 76 rating points13161240+76 rating
- CharXiv (RQ)Qwen3.5 27B by 2.177.479.5+2.1
- MMMU-Protie75.475tie
MathClaude Opus 4.6
- HMMT Feb. 2026Claude Opus 4.6 by 15.196.281.1+15.1
- AIME 2026Claude Opus 4.6 by 596.791.7+5
- IMOAnswerBenchQwen3.5 27B by 4.675.379.9+4.6
Following instructionsQwen3.5 27B
- Multi-ChallengeQwen3.5 27B by 23.637.160.8+23.6
- IFBenchQwen3.5 27B by 22.553.175.6+22.5
Other results22 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-OmniscienceClaude Opus 4.6 by 57.713.7-44+57.7
- BabyVisionQwen3.5 27B by 29.814.844.6+29.8
- Terminal-Bench 2.0Claude Opus 4.6 by 23.865.441.6+23.8
- NL2RepoClaude Opus 4.6 by 22.549.827.3+22.5
- MATH-VisionQwen3.5 27B by 20.565.586+20.5
- MCPMarkClaude Opus 4.6 by 20.456.736.3+20.4
- ERQAQwen3.5 27B by 19.740.860.5+19.7
- Tool-DecathlonClaude Opus 4.6 by 15.747.231.5+15.7
- Artificial Analysis Coding IndexClaude Opus 4.6 by 13.248.134.9+13.2
- AA Agentic IndexClaude Opus 4.6 by 1367.654.6+13
- ScreenSpot-Pro (No tools)Qwen3.5 27B by 12.657.770.3+12.6
- LVBenchQwen3.5 27B by 10.66373.6+10.6
- RealWorldQAQwen3.5 27B by 9.873.983.7+9.8
- AA IntelligenceClaude Opus 4.6 by 931.922.9+9
- HMMT Nov. 2025Claude Opus 4.6 by 6.596.389.8+6.5
- MMMLUClaude Opus 4.6 by 5.291.185.9+5.2
- vectara_avg_summary_lengthClaude Opus 4.6 by 43.2 rating points137.694.4+43.2 rating
- τ³-BenchClaude Opus 4.6 by 472.468.4+4
- MMLU-ProClaude Opus 4.6 by 389.186.1+3
- τ²-Bench Telecom (AA run)Qwen3.5 27B by 1.892.193.9+1.8
- vectara_factual_consistencytie87.887.9tie
- vectara_answer_ratetie99.899.8tie
Questions people ask
Which is better, Claude Opus 4.6 or Qwen3.5 27B?
Claude Opus 4.6 wins five of the eight areas we test: coding, agents, reasoning, facts and math. Qwen3.5 27B wins following instructions, and costs 91% less. They are level on images and charts and long documents.
Which is better for coding?
Claude Opus 4.6. It wins 7 of the 7 coding tests both models report; Qwen3.5 27B wins none.
Which is cheaper?
Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens; Qwen3.5 27B costs $0.30 and $2.40. That makes Qwen3.5 27B about 91% cheaper for the same work.
How do you compare the two?
We use the 48 benchmark tests both models have published scores on. The verdict counts the 26 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 22 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.