Claude Sonnet 4.6 vs Claude Opus 5
Wins 0 of 8 areas
—
Wins 6 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math
Claude Opus 5 is the stronger all-rounder.Claude Sonnet 4.6 is cheaper.
Scores updated · 43 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 software07Claude Opus 57 of 7 tests
- AgentsCarrying out multi-step tasks on its own06Claude Opus 56 of 6 tests
- ReasoningHard problems that need careful thinking05Claude Opus 55 of 5 tests
- Images and chartsUnderstanding pictures, charts and video03Claude Opus 53 of 3 tests
- FactsGetting facts right instead of making them up12Claude Opus 52 of 3 tests
- MathCompetition and research-level math01Claude Opus 51 of 1 test
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Math, 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.
Claude Sonnet 4.6 costs 40% 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 Claude Opus 5 pulls ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+46points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+32.1points ahead
- Short factual questions, answered correctlySimpleQA Verified+27.1points ahead
Where Claude Sonnet 4.6 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+12.4points ahead
Every test, side by side
All 43 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 5
- LiveBench · Agentic CodingClaude Opus 5 by 22.642.665.2+22.6
- SWE-bench ProClaude Opus 5 by 21.158.179.2+21.1
- Terminal-Bench 2.1Claude Opus 5 by 17.971.289.1+17.9
- LMArena · WebDevClaude Opus 5 by 170 rating points15221692+170 rating
- SWE-bench VerifiedClaude Opus 5 by 16.479.696+16.4
- SciCodeClaude Opus 5 by 6.350.156.4+6.3
- LiveBench · CodingClaude Opus 5 by 2.279.381.5+2.2
AgentsClaude Opus 5
- Terminal-Bench 4.0Claude Opus 5 by 46349+46
- GDPValClaude Opus 5 by 24.536.761.2+24.5
- MCP AtlasClaude Opus 5 by 24.561.385.8+24.5
- BrowseCompClaude Opus 5 by 16.174.790.8+16.1
- OSWorld-VerifiedClaude Opus 5 by 10.972.583.4+10.9
- τ-Bench V3 · BankingClaude Opus 5 by 7.734.442.1+7.7
ReasoningClaude Opus 5
- ARC-AGI-2Claude Opus 5 by 32.158.390.4+32.1
- CritPtClaude Opus 5 by 263.129.1+26
- Humanity's Last ExamClaude Opus 5 by 21.333.654.9+21.3
- LiveBench · ReasoningClaude Opus 5 by 6.484.891.2+6.4
- GPQA DiamondClaude Opus 5 by 5.787.593.2+5.7
FactsClaude Opus 5
- SimpleQA VerifiedClaude Opus 5 by 27.132.859.9+27.1
- AA-Omniscience · AccuracyClaude Opus 5 by 2040.960.9+20
- AA-Omniscience · Non-hallucinationClaude Sonnet 4.6 by 12.451.639.2+12.4
Images and chartsClaude Opus 5
- CharXiv (RQ)Claude Opus 5 by 12.171.683.7+12.1
- MMMU-ProClaude Opus 5 by 11.473.384.7+11.4
- LMArena · VisionClaude Opus 5 by 38 rating points12831321+38 rating
Following instructionsEven
- LiveBench · Instruction Followingtie63.263.8tie
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.
- GDPval-AA v2Claude Opus 5 by 429 rating points13951824+429 rating
- DeepSWE 1.1Claude Opus 5 by 38.83068.8+38.8
- AA-OmniscienceClaude Opus 5 by 24.912.237.1+24.9
- AA Agentic IndexClaude Opus 5 by 23.133.156.2+23.1
- AA IntelligenceClaude Opus 5 by 20.730.150.8+20.7
- HLE (with tools)Claude Opus 5 by 16.846.863.6+16.8
- HealthBench ProfessionalClaude Opus 5 by 15.644.259.8+15.6
- Artificial Analysis Coding IndexClaude Opus 5 by 156378+15
- OfficeQA ProClaude Opus 5 by 13.553.466.9+13.5
- livebench_languageClaude Opus 5 by 12.676.188.7+12.6
- ARC-AGI-1Claude Opus 5 by 11.58697.5+11.5
- OfficeQAClaude Opus 5 by 9.468.778.1+9.4
- GDP (Surge AI)Claude Opus 5 by 6.978.685.5+6.9
- Legal Agent BenchmarkClaude Opus 5 by 6.35.411.7+6.3
- livebench_data_analysisClaude Sonnet 4.6 by 3.47874.5+3.4
- GDPval-AA (Elo)Claude Opus 5 by 32 rating points16761708+32 rating
Questions people ask
Which is better, Claude Sonnet 4.6 or Claude Opus 5?
Claude Opus 5 wins six of the eight areas we test: coding, agents, reasoning, facts, images and charts and math. Claude Sonnet 4.6 wins none, but costs 40% less. They are level on long documents and following instructions.
Which is better for coding?
Claude Opus 5. It wins 7 of the 7 coding tests both models report; Claude Sonnet 4.6 wins none.
Which is cheaper?
Claude Sonnet 4.6 costs $3.00 per million input tokens and $15.00 per million output tokens; Claude Opus 5 costs $5.00 and $25.00. That makes Claude Sonnet 4.6 about 40% cheaper for the same work.
How do you compare the two?
We use the 43 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 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.