Claude Sonnet 4.5 vs Qwen3.5 122B A10B
Wins 3 of 7 areas
Coding · Agents · Facts
Wins 3 of 7 areas
Reasoning · Images and charts · Following instructions
The two are evenly matched.Claude Sonnet 4.5 is better at coding and agents; Qwen3.5 122B A10B at images and charts and reasoning.
Scores updated · 47 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.
- CodingWriting and fixing software62Claude Sonnet 4.56 of 8 tests
- AgentsCarrying out multi-step tasks on its own31Claude Sonnet 4.53 of 5 tests · 1 tie
- FactsGetting facts right instead of making them up20Claude Sonnet 4.52 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video04Qwen3.5 122B A10B4 of 4 tests
- ReasoningHard problems that need careful thinking02Qwen3.5 122B A10B2 of 3 tests · 1 tie
- Following instructionsDoing exactly what it is asked02Qwen3.5 122B A10B2 of 2 tests
- Long documentsFinding answers in very long texts11Even1 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.
Qwen3.5 122B A10B costs 80% 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 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+38points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+9.2points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+8.5points ahead
Where Qwen3.5 122B A10B pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+39.7points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+18.4points ahead
- Recent programming contest problemsLiveCodeBench v6+14.9points ahead
Every test, side by side
All 47 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 4.5
- LiveCodeBench v6Qwen3.5 122B A10B by 14.96478.9+14.9
- Terminal-Bench 2.1Claude Sonnet 4.5 by 8.255.847.6+8.2
- SciCodeClaude Sonnet 4.5 by 645.739.7+6
- SWE-bench VerifiedClaude Sonnet 4.5 by 5.277.272+5.2
- SWE-bench MultilingualClaude Sonnet 4.5 by 56762+5
- Terminal-Bench HardClaude Sonnet 4.5 by 4.535.631.1+4.5
- LMArena · WebDevClaude Sonnet 4.5 by 33 rating points13931360+33 rating
- SWE-bench ProQwen3.5 122B A10B by 1.343.644.9+1.3
AgentsClaude Sonnet 4.5
- BrowseCompQwen3.5 122B A10B by 39.724.163.8+39.7
- τ-Bench V3 · BankingClaude Sonnet 4.5 by 9.224.515.3+9.2
- GDPValClaude Sonnet 4.5 by 4.820.515.7+4.8
- OSWorld-VerifiedClaude Sonnet 4.5 by 3.461.458+3.4
- Terminal-Bench 4.0tie00tie
ReasoningQwen3.5 122B A10B
- Humanity's Last ExamQwen3.5 122B A10B by 7.417.825.2+7.4
- GPQA DiamondQwen3.5 122B A10B by 2.383.485.7+2.3
- CritPttie1.10.6tie
FactsClaude Sonnet 4.5
- AA-Omniscience · Non-hallucinationClaude Sonnet 4.5 by 3850.912.9+38
- AA-Omniscience · AccuracyClaude Sonnet 4.5 by 8.532.924.4+8.5
- Vectara HHEM hallucination ratelower is bettertie1211.2tie
Images and chartsQwen3.5 122B A10B
- CharXiv (RQ)Qwen3.5 122B A10B by 1067.277.2+10
- MathVistaQwen3.5 122B A10B by 7.679.887.4+7.6
- MMMU-ProQwen3.5 122B A10B by 6.368.775+6.3
- MMMUQwen3.5 122B A10B by 4.379.683.9+4.3
Long documentsEven
- AA-LCRQwen3.5 122B A10B by 472.376.3+4
- LongBench v2Claude Sonnet 4.5 by 1.661.860.2+1.6
Following instructionsQwen3.5 122B A10B
- IFBenchQwen3.5 122B A10B by 18.457.375.7+18.4
- Multi-ChallengeQwen3.5 122B A10B by 6.255.361.5+6.2
Other results20 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 Sonnet 4.5 by 41.4-0.1-41.5+41.4
- BrowseComp-ZHQwen3.5 122B A10B by 27.542.469.9+27.5
- HMMT 2025Qwen3.5 122B A10B by 15.774.690.3+15.7
- τ²-Bench Telecom (AA run)Qwen3.5 122B A10B by 15.578.193.6+15.5
- RealWorldQAQwen3.5 122B A10B by 14.870.385.1+14.8
- CC-OCRQwen3.5 122B A10B by 13.768.181.8+13.7
- AA Agentic IndexClaude Sonnet 4.5 by 7.917.59.6+7.9
- HallusionBenchQwen3.5 122B A10B by 7.759.967.6+7.7
- Artificial Analysis Coding IndexClaude Sonnet 4.5 by 6.452.145.7+6.4
- AI2DQwen3.5 122B A10B by 6.38793.3+6.3
- AA IntelligenceClaude Sonnet 4.5 by 5.120.715.6+5.1
- vectara_answer_rateQwen3.5 122B A10B by 4.295.699.8+4.2
- vectara_avg_summary_lengthClaude Sonnet 4.5 by 41.4 rating points127.886.4+41.4 rating
- SimpleVQAQwen3.5 122B A10B by 4.157.661.7+4.1
- OmniDocBench 1.5Qwen3.5 122B A10B by 485.889.8+4
- MMMLUClaude Sonnet 4.5 by 2.489.186.7+2.4
- MMLU-ReduxClaude Sonnet 4.5 by 1.695.694+1.6
- MMLU-Protie87.586.7tie
- vectara_factual_consistencytie8888.8tie
- Terminal-Bench 2.0tie5049.4tie
Questions people ask
Which is better, Claude Sonnet 4.5 or Qwen3.5 122B A10B?
Claude Sonnet 4.5 and Qwen3.5 122B A10B each win three of the seven areas where both have results. Claude Sonnet 4.5 wins coding, agents and facts; Qwen3.5 122B A10B wins reasoning, images and charts and following instructions. Qwen3.5 122B A10B costs 80% less. They are level on long documents.
Which is better for coding?
Claude Sonnet 4.5. It wins 6 of the 8 coding tests both models report; Qwen3.5 122B A10B wins 2.
Which is cheaper?
Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens; Qwen3.5 122B A10B costs $0.40 and $3.20. That makes Qwen3.5 122B A10B about 80% cheaper for the same work.
How do you compare the two?
We use the 47 benchmark tests both models have published scores on. The verdict counts the 27 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.