Claude Opus 4.7 vs Claude Sonnet 4.6
Wins 7 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Following instructions
Wins 1 of 8 areas
Long documents
Claude Opus 4.7 is the stronger all-rounder.Claude Sonnet 4.6 is cheaper and better at long documents.
Scores updated · 61 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 software71Claude Opus 4.77 of 8 tests
- AgentsCarrying out multi-step tasks on its own60Claude Opus 4.76 of 7 tests · 1 tie
- ReasoningHard problems that need careful thinking50Claude Opus 4.75 of 5 tests
- Images and chartsUnderstanding pictures, charts and video30Claude Opus 4.73 of 3 tests
- FactsGetting facts right instead of making them up31Claude Opus 4.73 of 4 tests
- MathCompetition and research-level math20Claude Opus 4.72 of 2 tests
- Following instructionsDoing exactly what it is asked20Claude Opus 4.72 of 2 tests
- Long documentsFinding answers in very long texts02Claude Sonnet 4.62 of 2 tests
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 three tests each model wins by the widest margin. Scores are out of 100.
Where Claude Opus 4.7 pulls ahead
- Reasoning about charts from research papersCharXiv (RQ)+19.4points ahead
- Short factual questions, answered correctlySimpleQA Verified+18.9points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+17.5points ahead
Where Claude Sonnet 4.6 pulls ahead
- Finds one of eight look-alike replies in a long chatMRCR v2 8 needle 128k (average)+25.6points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+1.5points ahead
- Reasons across sets of long documentsAA-LCR+1.3points ahead
Every test, side by side
All 61 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Opus 4.7
- Terminal-Bench 2.1Claude Opus 4.7 by 11.983.171.2+11.9
- LiveBench · Agentic CodingClaude Opus 4.7 by 8.150.742.6+8.1
- SWE-bench VerifiedClaude Opus 4.7 by 887.679.6+8
- SWE-bench ProClaude Opus 4.7 by 6.264.358.1+6.2
- SciCodeClaude Opus 4.7 by 4.454.550.1+4.4
- LMArena · WebDevClaude Opus 4.7 by 36 rating points15571521+36 rating
- LiveBench · CodingClaude Opus 4.7 by 2.882.179.3+2.8
- Terminal-Bench HardClaude Sonnet 4.6 by 1.551.553+1.5
AgentsClaude Opus 4.7
- MCP AtlasClaude Opus 4.7 by 1677.361.3+16
- AA IT-Bench SREClaude Opus 4.7 by 6.946.739.8+6.9
- GDPValClaude Opus 4.7 by 6.142.836.7+6.1
- AA ApexAgentsClaude Opus 4.7 by 5.933.928+5.9
- OSWorld-VerifiedClaude Opus 4.7 by 5.57872.5+5.5
- BrowseCompClaude Opus 4.7 by 4.679.374.7+4.6
- τ-Bench V3 · Bankingtie34.634.4tie
ReasoningClaude Opus 4.7
- ARC-AGI-2Claude Opus 4.7 by 17.575.858.3+17.5
- CritPtClaude Opus 4.7 by 8.9123.1+8.9
- Humanity's Last ExamClaude Opus 4.7 by 8.742.333.6+8.7
- GPQA DiamondClaude Opus 4.7 by 3.991.487.5+3.9
- LiveBench · ReasoningClaude Opus 4.7 by 2.487.284.8+2.4
FactsClaude Opus 4.7
- SimpleQA VerifiedClaude Opus 4.7 by 18.951.732.8+18.9
- AA-Omniscience · AccuracyClaude Opus 4.7 by 848.940.9+8
- AA-Omniscience · Non-hallucinationClaude Opus 4.7 by 6.157.751.6+6.1
- Vectara HHEM hallucination ratelower is betterClaude Sonnet 4.6 by 1.41210.6+1.4
Images and chartsClaude Opus 4.7
- CharXiv (RQ)Claude Opus 4.7 by 19.49171.6+19.4
- MMMU-ProClaude Opus 4.7 by 5.578.873.3+5.5
- LMArena · VisionClaude Opus 4.7 by 30 rating points13131283+30 rating
MathClaude Opus 4.7
- USAMO 2026Claude Opus 4.7 by 14.369.355+14.3
- LiveBench · MathematicsClaude Opus 4.7 by 5.992.887+5.9
Long documentsClaude Sonnet 4.6
- MRCR v2 8 needle 128k (average)Claude Sonnet 4.6 by 25.659.384.9+25.6
- AA-LCRClaude Sonnet 4.6 by 1.378.780+1.3
Following instructionsClaude Opus 4.7
- LiveBench · Instruction FollowingClaude Opus 4.7 by 3.566.763.2+3.5
- IFBenchClaude Opus 4.7 by 258.656.6+2
Other results28 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.
- OfficeQA ProClaude Opus 4.7 by 27.280.653.4+27.2
- ToolathlonClaude Opus 4.7 by 18.359.341+18.3
- Blueprint-Bench 2Claude Opus 4.7 by 17.824.56.7+17.8
- OfficeQAClaude Opus 4.7 by 17.686.368.7+17.6
- AA-OmniscienceClaude Opus 4.7 by 15.127.312.2+15.1
- Toolathlon Pass@∞Claude Opus 4.7 by 14.852.838+14.8
- frontiermath_tier_4_v1Claude Opus 4.7 by 14.622.98.3+14.6
- Humanity’s Last ExamClaude Opus 4.7 by 13.746.933.2+13.7
- τ²-Bench Telecom (AA run)Claude Opus 4.7 by 12.988.675.7+12.9
- AA IntelligenceClaude Opus 4.7 by 10.640.730.1+10.6
- Artificial Analysis Coding IndexClaude Opus 4.7 by 10.673.663+10.6
- Terminal-Bench 2.0Claude Opus 4.7 by 10.369.459.1+10.3
- Toolathlon Avg turnsClaude Opus 4.7 by 9.425.916.5+9.4
- HLE (with tools)Claude Opus 4.7 by 7.954.746.8+7.9
- GDPval-AA (Elo)Claude Opus 4.7 by 77 rating points17531676+77 rating
- AA Agentic IndexClaude Opus 4.7 by 6.439.533.1+6.4
- ARC-AGI-1Claude Opus 4.7 by 69286+6
- CharXiv Reasoning (With tools)Claude Opus 4.7 by 5.79185.3+5.7
- vectara_avg_summary_lengthClaude Opus 4.7 by 34.4 rating points149.1114.7+34.4 rating
- MMMLUClaude Opus 4.7 by 2.291.589.3+2.2
- vectara_answer_rateClaude Sonnet 4.6 by 1.99899.9+1.9
- livebench_languageClaude Opus 4.7 by 1.877.976.1+1.8
- Legal Agent BenchmarkClaude Opus 4.7 by 1.77.15.4+1.7
- LiveBenchClaude Opus 4.7 by 1.476.975.5+1.4
- vectara_factual_consistencyClaude Sonnet 4.6 by 1.48889.4+1.4
- Finance AgentClaude Opus 4.7 by 1.164.463.3+1.1
- Finance Agent v2tie51.551tie
- livebench_data_analysistie78.378tie
Questions people ask
Which is better, Claude Opus 4.7 or Claude Sonnet 4.6?
Claude Opus 4.7 wins seven of the eight areas we test: coding, agents, reasoning, facts, images and charts, math and following instructions. Claude Sonnet 4.6 wins long documents, and costs 40% less.
Which is better for coding?
Claude Opus 4.7. It wins 7 of the 8 coding tests both models report; Claude Sonnet 4.6 wins 1.
Which is cheaper?
Claude Opus 4.7 costs $5.00 per million input tokens and $25.00 per million output tokens; Claude Sonnet 4.6 costs $3.00 and $15.00. That makes Claude Sonnet 4.6 about 40% cheaper for the same work.
How do you compare the two?
We use the 61 benchmark tests both models have published scores on. The verdict counts the 33 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 28 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.