Claude Opus 4.6 vs Qwen3.8 Max Preview
Wins 0 of 8 areas
—
Wins 7 of 8 areas
Coding · Agents · Reasoning · Images and charts · Math · Long documents · Following instructions
Qwen3.8 Max Preview is the stronger all-rounder.
Scores updated · 39 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 thinking03Qwen3.8 Max Preview3 of 4 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own03Qwen3.8 Max Preview3 of 3 tests
- MathCompetition and research-level math03Qwen3.8 Max Preview3 of 3 tests
- CodingWriting and fixing software13Qwen3.8 Max Preview3 of 5 tests · 1 tie
- Following instructionsDoing exactly what it is asked02Qwen3.8 Max Preview2 of 2 tests
- Images and chartsUnderstanding pictures, charts and video01Qwen3.8 Max Preview1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts01Qwen3.8 Max Preview1 of 1 test
- FactsGetting facts right instead of making them up11Even1 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.
Qwen3.8 Max Preview costs 73% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Qwen3.8 Max Preview pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+34points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+29.7points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+19.5points ahead
Where Claude Opus 4.6 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+15.3points ahead
- Programming problems, refreshed regularlyLiveBench · Coding+5.3points ahead
Every test, side by side
All 39 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.8 Max Preview
- LiveBench · Agentic CodingQwen3.8 Max Preview by 15.74964.7+15.7
- SWE-bench ProQwen3.8 Max Preview by 14.353.467.7+14.3
- LMArena · WebDevQwen3.8 Max Preview by 125 rating points15471672+125 rating
- LiveBench · CodingClaude Opus 4.6 by 5.378.272.9+5.3
- SciCodetie51.952.1tie
AgentsQwen3.8 Max Preview
- OSWorld-VerifiedQwen3.8 Max Preview by 13.472.786.1+13.4
- AA ApexAgentsQwen3.8 Max Preview by 9.43342.4+9.4
- GDPValQwen3.8 Max Preview by 2.455.958.3+2.4
ReasoningQwen3.8 Max Preview
- CritPtQwen3.8 Max Preview by 5.112.617.7+5.1
- GPQA DiamondQwen3.8 Max Preview by 3.289.692.8+3.2
- Humanity's Last ExamQwen3.8 Max Preview by 3.239.943.1+3.2
- LiveBench · Reasoningtie88.788.2tie
FactsEven
- AA-Omniscience · Non-hallucinationQwen3.8 Max Preview by 3437.271.2+34
- AA-Omniscience · AccuracyClaude Opus 4.6 by 15.34731.7+15.3
- SimpleQA Verifiedtie4747.3tie
Images and chartsQwen3.8 Max Preview
- MMMU-ProQwen3.8 Max Preview by 7.475.482.8+7.4
- LMArena · Visiontie13161314tie
MathQwen3.8 Max Preview
- FrontierMath Tier 4Qwen3.8 Max Preview by 19.526.846.3+19.5
- FrontierMath Tiers 1-3 (v2)Qwen3.8 Max Preview by 8.76674.7+8.7
- LiveBench · MathematicsQwen3.8 Max Preview by 289.391.3+2
Long documentsQwen3.8 Max Preview
- AA-LCRQwen3.8 Max Preview by 2.37880.3+2.3
Following instructionsQwen3.8 Max Preview
- IFBenchQwen3.8 Max Preview by 29.753.182.8+29.7
- LiveBench · Instruction FollowingQwen3.8 Max Preview by 10.863.374.1+10.8
Other results16 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.
- ERQAQwen3.8 Max Preview by 3740.877.8+37
- ScreenSpot-Pro (No tools)Qwen3.8 Max Preview by 26.857.784.5+26.8
- Artificial Analysis Coding IndexQwen3.8 Max Preview by 23.748.171.8+23.7
- AndroidWorldQwen3.8 Max Preview by 23.36285.3+23.3
- LVBenchQwen3.8 Max Preview by 18.86381.8+18.8
- AA Agentic IndexClaude Opus 4.6 by 1867.649.6+18
- JobBenchQwen3.8 Max Preview by 16.836.653.4+16.8
- ToolathlonQwen3.8 Max Preview by 15.756.872.5+15.7
- RealWorldQAQwen3.8 Max Preview by 14.173.988+14.1
- AA IntelligenceQwen3.8 Max Preview by 13.531.945.4+13.5
- livebench_data_analysisQwen3.8 Max Preview by 8.569.978.4+8.5
- NL2RepoQwen3.8 Max Preview by 6.149.855.9+6.1
- CyberGymQwen3.8 Max Preview by 4.773.878.5+4.7
- livebench_languageClaude Opus 4.6 by 3.683.379.7+3.6
- HLE (with tools)Qwen3.8 Max Preview by 2.953.356.2+2.9
- AA-OmniscienceClaude Opus 4.6 by 1.713.712+1.7
Questions people ask
Which is better, Claude Opus 4.6 or Qwen3.8 Max Preview?
Qwen3.8 Max Preview wins seven of the eight areas we test: coding, agents, reasoning, images and charts, math, long documents and following instructions. Claude Opus 4.6 wins none. They are level on facts.
Which is better for coding?
Qwen3.8 Max Preview. It wins 3 of the 5 coding tests both models report; Claude Opus 4.6 wins 1, and 1 is a tie.
Which is cheaper?
Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens; Qwen3.8 Max Preview costs $2.00 and $6.00. That makes Qwen3.8 Max Preview about 73% cheaper for the same work.
How do you compare the two?
We use the 39 benchmark tests both models have published scores on. The verdict counts the 23 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 16 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.