Claude Sonnet 4 vs Qwen3.5 35B A3B
Wins 1 of 7 areas
Facts
Wins 5 of 7 areas
Coding · Reasoning · Images and charts · Long documents · Following instructions
Qwen3.5 35B A3B is the stronger all-rounder.Claude Sonnet 4 is better at facts.
Scores updated · 39 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. 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 thinking02Qwen3.5 35B A3B2 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video02Qwen3.5 35B A3B2 of 2 tests
- Following instructionsDoing exactly what it is asked02Qwen3.5 35B A3B2 of 2 tests
- CodingWriting and fixing software34Qwen3.5 35B A3B4 of 7 tests
- Long documentsFinding answers in very long texts01Qwen3.5 35B A3B1 of 1 test
- FactsGetting facts right instead of making them up20Claude Sonnet 42 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own22Even2 each
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 35B A3B 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 Qwen3.5 35B A3B pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+48.8points ahead
- Recent programming contest problemsLiveCodeBench v6+26.1points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+17.8points ahead
Where Claude Sonnet 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+56.3points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+11.8points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+4.6points 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.
CodingQwen3.5 35B A3B
- LiveCodeBench v6Qwen3.5 35B A3B by 26.148.574.6+26.1
- Terminal-Bench HardClaude Sonnet 4 by 4.631.126.5+4.6
- Terminal-Bench 2.1Qwen3.5 35B A3B by 4.536.340.8+4.5
- SWE-bench VerifiedClaude Sonnet 4 by 3.572.769.2+3.5
- SWE-bench MultilingualQwen3.5 35B A3B by 3.456.960.3+3.4
- SciCodeClaude Sonnet 4 by 2.34037.7+2.3
- SWE-bench ProQwen3.5 35B A3B by 1.942.744.6+1.9
AgentsEven
- BrowseCompQwen3.5 35B A3B by 48.812.261+48.8
- OSWorld-VerifiedQwen3.5 35B A3B by 12.342.254.5+12.3
- τ-Bench V3 · BankingClaude Sonnet 4 by 11.816.74.9+11.8
- GDPValClaude Sonnet 4 by 4.19.75.6+4.1
ReasoningQwen3.5 35B A3B
- Humanity's Last ExamQwen3.5 35B A3B by 10.310.721+10.3
- GPQA DiamondQwen3.5 35B A3B by 6.877.784.5+6.8
- CritPttie0.30.9tie
FactsClaude Sonnet 4
- AA-Omniscience · Non-hallucinationClaude Sonnet 4 by 56.370.914.6+56.3
- AA-Omniscience · AccuracyClaude Sonnet 4 by 2.622.720.1+2.6
- Vectara HHEM hallucination ratelower is bettertie10.310.5tie
Images and chartsQwen3.5 35B A3B
Following instructionsQwen3.5 35B A3B
- IFBenchQwen3.5 35B A3B by 17.854.772.5+17.8
- Multi-ChallengeQwen3.5 35B A3B by 2.957.160+2.9
Other results17 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.
- HMMT 2025Qwen3.5 35B A3B by 73.315.989.2+73.3
- AA-OmniscienceClaude Sonnet 4 by 48.30.2-48.1+48.3
- BrowseComp-ZHQwen3.5 35B A3B by 40.429.169.5+40.4
- τ²-Bench Telecom (AA run)Qwen3.5 35B A3B by 24.664.689.2+24.6
- OJBenchQwen3.5 35B A3B by 20.715.336+20.7
- MMLUClaude Sonnet 4 by 10.491.581.1+10.4
- SuperGPQAQwen3.5 35B A3B by 7.755.763.4+7.7
- AA Agentic IndexClaude Sonnet 4 by 5.817.611.8+5.8
- vectara_avg_summary_lengthClaude Sonnet 4 by 50.9 rating points145.894.9+50.9 rating
- IFEvalQwen3.5 35B A3B by 4.387.691.9+4.3
- MMLU-ProQwen3.5 35B A3B by 1.683.785.3+1.6
- MMMLUClaude Sonnet 4 by 1.386.585.2+1.3
- vectara_answer_rateQwen3.5 35B A3B by 1.298.699.8+1.2
- Artificial Analysis Coding Indextie37.637tie
- AA Intelligencetie18.919.3tie
- MMLU-Reduxtie93.693.3tie
- vectara_factual_consistencytie89.789.5tie
Questions people ask
Which is better, Claude Sonnet 4 or Qwen3.5 35B A3B?
Qwen3.5 35B A3B wins five of the seven areas where both have results: coding, reasoning, images and charts, long documents and following instructions. Claude Sonnet 4 wins facts. They are level on agents.
Which is better for coding?
Qwen3.5 35B A3B. It wins 4 of the 7 coding tests both models report; Claude Sonnet 4 wins 3.
Which is cheaper?
Claude Sonnet 4 costs $3.00 per million input tokens and $15.00 per million output tokens; Qwen3.5 35B A3B costs $0.25 and $2.00. That makes Qwen3.5 35B A3B about 88% 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 22 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 17 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.