Claude Opus 4.7 vs Qwen3.8 27B
Wins 4 of 8 areas
Coding · Reasoning · Images and charts · Math
Wins 3 of 8 areas
Agents · Long documents · Following instructions
Claude Opus 4.7 wins more areas, narrowly.Qwen3.8 27B is cheaper and better at agents.
Scores updated · 41 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 thinking50Claude Opus 4.75 of 6 tests · 1 tie
- CodingWriting and fixing software62Claude Opus 4.76 of 8 tests
- Images and chartsUnderstanding pictures, charts and video21Claude Opus 4.72 of 4 tests · 1 tie
- MathCompetition and research-level math10Claude Opus 4.71 of 1 test
- AgentsCarrying out multi-step tasks on its own13Qwen3.8 27B3 of 5 tests · 1 tie
- Following instructionsDoing exactly what it is asked02Qwen3.8 27B2 of 2 tests
- Long documentsFinding answers in very long texts01Qwen3.8 27B1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Math 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.
Qwen3.8 27B 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 4.7 pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+33.4points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+33.3points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+17.7points ahead
Where Qwen3.8 27B pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+20.9points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+13.4points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+12points ahead
Every test, side by side
All 41 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4.7
- LiveBench · Agentic CodingQwen3.8 27B by 10.750.761.4+10.7
- SWE-bench VerifiedClaude Opus 4.7 by 9.887.677.8+9.8
- SciCodeClaude Opus 4.7 by 7.954.546.6+7.9
- SWE-bench MultilingualClaude Opus 4.7 by 6.780.573.8+6.7
- LiveBench · CodingClaude Opus 4.7 by 6.482.175.7+6.4
- LMArena · WebDevQwen3.8 27B by 33 rating points15571590+33 rating
- Terminal-Bench 2.1Claude Opus 4.7 by 3.383.179.8+3.3
- SWE-bench ProClaude Opus 4.7 by 2.664.361.7+2.6
AgentsQwen3.8 27B
- BrowseCompClaude Opus 4.7 by 17.779.361.6+17.7
- τ-Bench V3 · BankingQwen3.8 27B by 13.434.648+13.4
- OSWorld-VerifiedQwen3.8 27B by 6.37884.3+6.3
- GDPValQwen3.8 27B by 3.442.646+3.4
- MCP Atlastie77.377.3tie
ReasoningClaude Opus 4.7
- ARC-AGI-2Claude Opus 4.7 by 33.475.842.4+33.4
- Humanity's Last ExamClaude Opus 4.7 by 8.442.333.9+8.4
- LiveBench · ReasoningClaude Opus 4.7 by 7.287.280+7.2
- CritPtClaude Opus 4.7 by 6.6125.4+6.6
- SimpleBenchClaude Opus 4.7 by 1.561.760.2+1.5
- GPQA Diamondtie91.490.5tie
FactsEven
- AA-Omniscience · AccuracyClaude Opus 4.7 by 33.348.915.6+33.3
- AA-Omniscience · Non-hallucinationQwen3.8 27B by 1257.769.7+12
Images and chartsClaude Opus 4.7
- OCRBenchv2Qwen3.8 27B by 6.656.963.5+6.6
- LMArena · VisionClaude Opus 4.7 by 43 rating points13141271+43 rating
- MMMU-ProClaude Opus 4.7 by 2.578.876.3+2.5
- CharXiv (RQ)tie9190.2tie
Following instructionsQwen3.8 27B
- IFBenchQwen3.8 27B by 20.958.679.5+20.9
- LiveBench · Instruction FollowingQwen3.8 27B by 666.772.7+6
Other results12 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.
- BabyVisionQwen3.8 27B by 63.422.285.6+63.4
- AA-OmniscienceClaude Opus 4.7 by 37.327.3-10+37.3
- GraphWalks BFS 256KClaude Opus 4.7 by 26.576.950.4+26.5
- ERQAQwen3.8 27B by 1352.565.5+13
- RealWorldQAQwen3.8 27B by 10.375.685.9+10.3
- AA Agentic IndexQwen3.8 27B by 739.546.5+7
- AA IntelligenceClaude Opus 4.7 by 740.733.7+7
- Artificial Analysis Coding IndexClaude Opus 4.7 by 5.573.668.1+5.5
- ARC-AGI-1Claude Opus 4.7 by 4.59287.5+4.5
- SWE-bench MultimodalQwen3.8 27B by 4.134.538.6+4.1
- livebench_languageClaude Opus 4.7 by 3.577.974.3+3.5
- livebench_data_analysisClaude Opus 4.7 by 1.778.376.6+1.7
Questions people ask
Which is better, Claude Opus 4.7 or Qwen3.8 27B?
Claude Opus 4.7 wins four of the eight areas we test: coding, reasoning, images and charts and math. Qwen3.8 27B wins agents, long documents and following instructions, and costs 88% less. They are level on facts.
Which is better for coding?
Claude Opus 4.7. It wins 6 of the 8 coding tests both models report; Qwen3.8 27B wins 2.
Which is cheaper?
Claude Opus 4.7 costs $5.00 per million input tokens and $25.00 per million output tokens; Qwen3.8 27B costs $0.50 and $3.00. That makes Qwen3.8 27B about 88% cheaper for the same work.
How do you compare the two?
We use the 41 benchmark tests both models have published scores on. The verdict counts the 29 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 12 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.