Qwen3.8 Max Preview vs Claude Sonnet 5.5
Wins 2 of 8 areas
Images and charts · Following instructions
Wins 5 of 8 areas
Coding · Agents · Reasoning · Math · Long documents
Claude Sonnet 5.5 is the stronger all-rounder.Qwen3.8 Max Preview is cheaper and better at images and charts.
Scores updated · 31 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 software14Claude Sonnet 5.54 of 5 tests
- ReasoningHard problems that need careful thinking03Claude Sonnet 5.53 of 3 tests
- MathCompetition and research-level math03Claude Sonnet 5.53 of 3 tests
- AgentsCarrying out multi-step tasks on its own02Claude Sonnet 5.52 of 2 tests
- Long documentsFinding answers in very long texts01Claude Sonnet 5.51 of 1 test
- Images and chartsUnderstanding pictures, charts and video10Qwen3.8 Max Preview1 of 1 test
- Following instructionsDoing exactly what it is asked10Qwen3.8 Max Preview1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each · 1 tie
Images and charts, long documents and following instructions 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 Max Preview costs 33% 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 Sonnet 5.5 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+34.2points ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+24.7points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+22.3points ahead
Where Qwen3.8 Max Preview pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+18.2points ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+17.3points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+8.4points ahead
Every test, side by side
All 31 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 5.5
- LiveBench · CodingClaude Sonnet 5.5 by 18.572.991.4+18.5
- SWE-bench ProClaude Sonnet 5.5 by 13.667.781.3+13.6
- LMArena · WebDevClaude Sonnet 5.5 by 104 rating points16701774+104 rating
- SciCodeClaude Sonnet 5.5 by 8.952.161+8.9
- LiveBench · Agentic CodingQwen3.8 Max Preview by 8.464.756.3+8.4
AgentsClaude Sonnet 5.5
- Terminal-Bench 4.0Claude Sonnet 5.5 by 24.738.963.6+24.7
- GDPValClaude Sonnet 5.5 by 8.658.366.9+8.6
ReasoningClaude Sonnet 5.5
- CritPtClaude Sonnet 5.5 by 13.717.731.4+13.7
- Humanity's Last ExamClaude Sonnet 5.5 by 11.943.155+11.9
- LiveBench · ReasoningClaude Sonnet 5.5 by 3.488.291.6+3.4
FactsEven
- AA-Omniscience · AccuracyClaude Sonnet 5.5 by 22.331.754+22.3
- AA-Omniscience · Non-hallucinationQwen3.8 Max Preview by 18.271.253+18.2
- SimpleQA Verifiedtie47.346.5tie
Images and chartsQwen3.8 Max Preview
- LMArena · VisionQwen3.8 Max Preview by 28 rating points13121284+28 rating
MathClaude Sonnet 5.5
- FrontierMath Tier 4Claude Sonnet 5.5 by 34.246.380.5+34.2
- FrontierMath Tiers 1-3 (v2)Claude Sonnet 5.5 by 14.174.788.8+14.1
- LiveBench · MathematicsClaude Sonnet 5.5 by 4.891.396.1+4.8
Long documentsClaude Sonnet 5.5
- AA-LCRClaude Sonnet 5.5 by 2.480.382.7+2.4
Following instructionsQwen3.8 Max Preview
- LiveBench · Instruction FollowingQwen3.8 Max Preview by 17.374.156.8+17.3
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.
- Terminal-Bench-Science 0.1Claude Sonnet 5.5 by 41.411.953.3+41.4
- CritPt v2 HardClaude Sonnet 5.5 by 29.529.559+29.5
- HLE-DiamondClaude Sonnet 5.5 by 29.32453.3+29.3
- AA-OmniscienceClaude Sonnet 5.5 by 20.31232.3+20.3
- livebench_data_analysisQwen3.8 Max Preview by 18.978.459.5+18.9
- DeepSWE 1.1Claude Sonnet 5.5 by 14.456.671+14.4
- AA IntelligenceClaude Sonnet 5.5 by 10.645.456+10.6
- HLE (with tools)Claude Sonnet 5.5 by 8.356.264.5+8.3
- Toolathlon VerifiedClaude Sonnet 5.5 by 5.372.577.8+5.3
- GDP (Surge AI)Claude Sonnet 5.5 by 322.825.8+3
- HealthBenchClaude Sonnet 5.5 by 1.960.262.1+1.9
- livebench_languageQwen3.8 Max Preview by 1.779.778+1.7
Questions people ask
Which is better, Qwen3.8 Max Preview or Claude Sonnet 5.5?
Claude Sonnet 5.5 wins five of the eight areas we test: coding, agents, reasoning, math and long documents. Qwen3.8 Max Preview wins images and charts and following instructions, and costs 33% less. They are level on facts.
Which is better for coding?
Claude Sonnet 5.5. It wins 4 of the 5 coding tests both models report; Qwen3.8 Max Preview wins 1.
Which is cheaper?
Qwen3.8 Max Preview costs $2.00 per million input tokens and $6.00 per million output tokens; Claude Sonnet 5.5 costs $2.00 and $10.00. That makes Qwen3.8 Max Preview about 33% cheaper for the same work.
How do you compare the two?
We use the 31 benchmark tests both models have published scores on. The verdict counts the 19 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.