Claude Opus 4.6 vs Claude Sonnet 5
Wins 1 of 8 areas
Facts
Wins 4 of 8 areas
Coding · Reasoning · Images and charts · Math
Claude Sonnet 5 wins more areas, narrowly.Claude Opus 4.6 is better at facts.
Scores updated · 45 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 software05Claude Sonnet 55 of 7 tests · 2 ties
- MathCompetition and research-level math03Claude Sonnet 53 of 4 tests · 1 tie
- ReasoningHard problems that need careful thinking13Claude Sonnet 53 of 5 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video12Claude Sonnet 52 of 3 tests
- FactsGetting facts right instead of making them up21Claude Opus 4.62 of 3 tests
- AgentsCarrying out multi-step tasks on its own11Even1 each · 1 tie
- Long documentsFinding answers in very long texts11Even1 each
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Following instructions 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.
Claude Sonnet 5 costs 60% 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 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+23.4points ahead
- Proof problems from the US Math OlympiadUSAMO 2026+13.3points ahead
- Reasoning about charts from research papersCharXiv (RQ)+10.9points ahead
Where Claude Opus 4.6 pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+13.3points ahead
- Real work tasks from 44 professionsGDPVal+7.6points ahead
- Common-sense trick questionsSimpleBench+7points ahead
Every test, side by side
All 45 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 5
- LiveBench · Agentic CodingClaude Sonnet 5 by 10.44959.4+10.4
- SWE-bench ProClaude Sonnet 5 by 9.853.463.2+9.8
- SWE-bench VerifiedClaude Sonnet 5 by 4.480.885.2+4.4
- LiveBench · CodingClaude Sonnet 5 by 2.578.280.7+2.5
- SciCodeClaude Sonnet 5 by 2.451.954.3+2.4
- LMArena · WebDevtie15461541tie
- SWE-bench Multilingualtie77.878.3tie
AgentsEven
- OSWorld-VerifiedClaude Sonnet 5 by 8.572.781.2+8.5
- GDPValClaude Opus 4.6 by 7.655.948.3+7.6
- BrowseComptie8484.7tie
ReasoningClaude Sonnet 5
- SimpleBenchClaude Opus 4.6 by 767.660.6+7
- CritPtClaude Sonnet 5 by 4.312.616.9+4.3
- GPQA DiamondClaude Sonnet 5 by 1.589.691.1+1.5
- Humanity's Last ExamClaude Sonnet 5 by 1.439.941.3+1.4
- LiveBench · Reasoningtie88.788.7tie
FactsClaude Opus 4.6
- AA-Omniscience · Non-hallucinationClaude Sonnet 5 by 23.437.260.6+23.4
- SimpleQA VerifiedClaude Opus 4.6 by 13.34733.7+13.3
- AA-Omniscience · AccuracyClaude Opus 4.6 by 6.94740+6.9
Images and chartsClaude Sonnet 5
- CharXiv (RQ)Claude Sonnet 5 by 10.977.488.3+10.9
- LMArena · VisionClaude Opus 4.6 by 41 rating points13161275+41 rating
- MMMU-ProClaude Sonnet 5 by 1.975.477.3+1.9
MathClaude Sonnet 5
- USAMO 2026Claude Sonnet 5 by 13.366.279.5+13.3
- LiveBench · MathematicsClaude Sonnet 5 by 3.689.392.9+3.6
- FrontierMath Tier 4Claude Sonnet 5 by 2.526.829.3+2.5
- FrontierMath Tiers 1-3 (v2)tie6665.6tie
Long documentsEven
- AA-LCRClaude Sonnet 5 by 47882+4
- MRCR v2 8 needle 128k (average)Claude Opus 4.6 by 2.58481.5+2.5
Following instructionsEven
- LiveBench · Instruction Followingtie63.363.9tie
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.
- Artificial Analysis Coding IndexClaude Sonnet 5 by 23.448.171.5+23.4
- AA Agentic IndexClaude Opus 4.6 by 23.367.644.3+23.3
- Terminal-Bench 2.0Claude Sonnet 5 by 1565.480.4+15
- livebench_languageClaude Opus 4.6 by 8.383.375+8.3
- Toolathlon Pass@∞Claude Opus 4.6 by 6.547.240.7+6.5
- AA IntelligenceClaude Sonnet 5 by 6.331.938.2+6.3
- LVBenchClaude Sonnet 5 by 5.56368.5+5.5
- HLE (with tools)Claude Sonnet 5 by 4.153.357.4+4.1
- CharXiv Reasoning (With tools)Claude Sonnet 5 by 3.684.788.3+3.6
- AA-OmniscienceClaude Sonnet 5 by 2.813.716.4+2.8
- ToolathlonClaude Opus 4.6 by 2.556.854.3+2.5
- OfficeQA ProClaude Sonnet 5 by 2.357.159.4+2.3
- livebench_data_analysisClaude Sonnet 5 by 1.869.971.7+1.8
- Legal Agent BenchmarkClaude Sonnet 5 by 1.64.25.8+1.6
- OfficeQAClaude Sonnet 5 by 1.673.575.1+1.6
- SWE-bench MultimodalClaude Sonnet 5 by 127.128.1+1
- BrowseComp (multi-agent)tie86.686.6tie
Questions people ask
Which is better, Claude Opus 4.6 or Claude Sonnet 5?
Claude Sonnet 5 wins four of the eight areas we test: coding, reasoning, images and charts and math. Claude Opus 4.6 wins facts. They are level on agents, long documents and following instructions.
Which is better for coding?
Claude Sonnet 5. It wins 5 of the 7 coding tests both models report; Claude Opus 4.6 wins none, and 2 are ties.
Which is cheaper?
Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens; Claude Sonnet 5 costs $2.00 and $10.00. That makes Claude Sonnet 5 about 60% cheaper for the same work.
How do you compare the two?
We use the 45 benchmark tests both models have published scores on. The verdict counts the 28 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.