Claude Opus 4.7 vs GPT-5.6 Sol
Wins 1 of 8 areas
Images and charts
Wins 7 of 8 areas
Coding · Agents · Reasoning · Facts · Math · Long documents · Following instructions
GPT-5.6 Sol is the stronger all-rounder.Claude Opus 4.7 is better at images and charts.
Both rank among the ten best models we track in coding.
Scores updated · 53 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.
- AgentsCarrying out multi-step tasks on its own07GPT-5.6 Sol7 of 7 tests
- ReasoningHard problems that need careful thinking07GPT-5.6 Sol7 of 7 tests
- CodingWriting and fixing software06GPT-5.6 Sol6 of 7 tests · 1 tie
- MathCompetition and research-level math04GPT-5.6 Sol4 of 4 tests
- Following instructionsDoing exactly what it is asked02GPT-5.6 Sol2 of 2 tests
- FactsGetting facts right instead of making them up12GPT-5.6 Sol2 of 4 tests · 1 tie
- Long documentsFinding answers in very long texts01GPT-5.6 Sol1 of 1 test
- Images and chartsUnderstanding pictures, charts and video21Claude Opus 4.72 of 3 tests
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.
GPT-5.6 Sol costs 20% 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 GPT-5.6 Sol pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+51.3points ahead
- Research-level physics problemsCritPt+20.3points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+18.9points ahead
Where Claude Opus 4.7 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+49.9points ahead
- Reasoning about charts from research papersCharXiv (RQ)+5.2points ahead
Every test, side by side
All 53 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.6 Sol
- Terminal-Bench HardGPT-5.6 Sol by 14.451.565.9+14.4
- LMArena · WebDevGPT-5.6 Sol by 60 rating points15581618+60 rating
- LiveBench · Agentic CodingGPT-5.6 Sol by 5.550.756.2+5.5
- Terminal-Bench 2.1GPT-5.6 Sol by 4.983.188+4.9
- SciCodeGPT-5.6 Sol by 2.654.557.1+2.6
- LiveBench · CodingGPT-5.6 Sol by 1.882.183.9+1.8
- SWE-bench Protie64.364.6tie
AgentsGPT-5.6 Sol
- GDPValGPT-5.6 Sol by 12.842.855.6+12.8
- BrowseCompGPT-5.6 Sol by 11.179.390.4+11.1
- τ-Bench V3 · BankingGPT-5.6 Sol by 9.734.644.3+9.7
- AA IT-Bench SREGPT-5.6 Sol by 9.546.756.2+9.5
- MCP AtlasGPT-5.6 Sol by 6.377.383.6+6.3
- AA ApexAgentsGPT-5.6 Sol by 633.939.9+6
- OSWorld-VerifiedGPT-5.6 Sol by 57883+5
ReasoningGPT-5.6 Sol
- CritPtGPT-5.6 Sol by 20.31232.3+20.3
- ARC-AGI-2GPT-5.6 Sol by 16.775.892.5+16.7
- ARC-AGI-3GPT-5.6 Sol by 7.60.27.8+7.6
- Humanity's Last ExamGPT-5.6 Sol by 7.242.349.5+7.2
- LiveBench · ReasoningGPT-5.6 Sol by 4.587.291.7+4.5
- SimpleBenchGPT-5.6 Sol by 3.161.764.8+3.1
- GPQA DiamondGPT-5.6 Sol by 2.791.494.1+2.7
FactsGPT-5.6 Sol
- AA-Omniscience · Non-hallucinationClaude Opus 4.7 by 49.957.77.8+49.9
- SimpleQA VerifiedGPT-5.6 Sol by 1851.769.7+18
- AA-Omniscience · AccuracyGPT-5.6 Sol by 10.548.959.4+10.5
- Vectara HHEM hallucination ratelower is bettertie1212.4tie
Images and chartsClaude Opus 4.7
- CharXiv (RQ)Claude Opus 4.7 by 5.29185.8+5.2
- MMMU-ProGPT-5.6 Sol by 4.678.883.4+4.6
- LMArena · VisionClaude Opus 4.7 by 34 rating points13131279+34 rating
MathGPT-5.6 Sol
- FrontierMath Tier 4GPT-5.6 Sol by 51.331.783+51.3
- FrontierMath Tiers 1-3 (v2)GPT-5.6 Sol by 18.970.289.1+18.9
- AIME 2026GPT-5.6 Sol by 4.195.899.9+4.1
- LiveBench · MathematicsGPT-5.6 Sol by 3.392.896.2+3.3
Following instructionsGPT-5.6 Sol
- IFBenchGPT-5.6 Sol by 14.158.672.7+14.1
- LiveBench · Instruction FollowingGPT-5.6 Sol by 5.266.771.8+5.2
Other results18 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.
- DeepSWEGPT-5.6 Sol by 18.75472.7+18.7
- OfficeQA ProClaude Opus 4.7 by 17.480.663.2+17.4
- CyberGymGPT-5.6 Sol by 11.473.184.5+11.4
- AA Agentic IndexGPT-5.6 Sol by 1139.550.5+11
- HLE (with tools)GPT-5.6 Sol by 9.854.764.5+9.8
- livebench_languageGPT-5.6 Sol by 9.877.987.7+9.8
- HiL-Bench (Tools-allowed)Claude Opus 4.7 by 9.441.732.3+9.4
- AA IntelligenceGPT-5.6 Sol by 6.340.747+6.3
- AA-OmniscienceClaude Opus 4.7 by 5.327.322+5.3
- ARC-AGI-1GPT-5.6 Sol by 4.59296.5+4.5
- Artificial Analysis Coding IndexGPT-5.6 Sol by 3.873.677.4+3.8
- τ²-Bench Telecom (AA run)Claude Opus 4.7 by 3.588.685.1+3.5
- DeepSWE v1.1 (Resolved)GPT-5.6 Sol by 3.269.873+3.2
- vectara_avg_summary_lengthClaude Opus 4.7 by 18 rating points149.1131.1+18 rating
- livebench_data_analysisGPT-5.6 Sol by 1.578.379.8+1.5
- ToolathlonClaude Opus 4.7 by 1.359.358+1.3
- vectara_answer_rateGPT-5.6 Sol by 1.19899.1+1.1
- vectara_factual_consistencytie8887.6tie
Questions people ask
Which is better, Claude Opus 4.7 or GPT-5.6 Sol?
GPT-5.6 Sol wins seven of the eight areas we test: coding, agents, reasoning, facts, math, long documents and following instructions. Claude Opus 4.7 wins images and charts.
Which is better for coding?
GPT-5.6 Sol. It wins 6 of the 7 coding tests both models report; Claude Opus 4.7 wins none, and 1 is a tie.
Which is cheaper?
Claude Opus 4.7 costs $5.00 per million input tokens and $25.00 per million output tokens; GPT-5.6 Sol costs $4.00 and $20.00. That makes GPT-5.6 Sol about 20% cheaper for the same work.
How do you compare the two?
We use the 53 benchmark tests both models have published scores on. The verdict counts the 35 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.