Claude Opus 4.5 vs GPT-5.5
Wins 0 of 8 areas
—
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
GPT-5.5 is the stronger all-rounder.Claude Opus 4.5 is cheaper.
Scores updated · 52 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 software06GPT-5.56 of 6 tests
- ReasoningHard problems that need careful thinking05GPT-5.55 of 5 tests
- MathCompetition and research-level math05GPT-5.55 of 5 tests
- AgentsCarrying out multi-step tasks on its own13GPT-5.53 of 4 tests
- FactsGetting facts right instead of making them up13GPT-5.53 of 4 tests
- Images and chartsUnderstanding pictures, charts and video02GPT-5.52 of 2 tests
- Following instructionsDoing exactly what it is asked02GPT-5.52 of 2 tests
- Long documentsFinding answers in very long texts01GPT-5.51 of 1 test
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.
Claude Opus 4.5 costs 14% 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-5.5 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+67.6points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+50.9points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+47.4points ahead
Where Claude Opus 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+28points ahead
- Real work tasks from 44 professionsGDPVal+4.6points ahead
Every test, side by side
All 52 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.5
- LiveBench · Agentic CodingGPT-5.5 by 14.339.754+14.3
- Terminal-Bench HardGPT-5.5 by 13.64760.6+13.6
- SWE-bench ProGPT-5.5 by 6.65258.6+6.6
- SciCodeGPT-5.5 by 6.349.555.8+6.3
- LMArena · WebDevGPT-5.5 by 44 rating points14681512+44 rating
- LiveBench · CodingGPT-5.5 by 2.579.782.2+2.5
AgentsGPT-5.5
- BrowseCompGPT-5.5 by 47.43784.4+47.4
- MCP AtlasGPT-5.5 by 1362.375.3+13
- OSWorld-VerifiedGPT-5.5 by 12.466.378.7+12.4
- GDPValClaude Opus 4.5 by 4.647.342.7+4.6
ReasoningGPT-5.5
- ARC-AGI-2GPT-5.5 by 47.437.685+47.4
- CritPtGPT-5.5 by 22.54.627.1+22.5
- Humanity's Last ExamGPT-5.5 by 15.730.145.8+15.7
- LiveBench · ReasoningGPT-5.5 by 9.680.189.7+9.6
- GPQA DiamondGPT-5.5 by 6.986.693.5+6.9
FactsGPT-5.5
- AA-Omniscience · Non-hallucinationClaude Opus 4.5 by 283911+28
- SimpleQA VerifiedGPT-5.5 by 17.345.763+17.3
- AA-Omniscience · AccuracyGPT-5.5 by 11.446.658+11.4
- Vectara HHEM hallucination ratelower is betterGPT-5.5 by 1.610.99.3+1.6
Images and chartsGPT-5.5
MathGPT-5.5
- FrontierMath Tier 4GPT-5.5 by 67.64.972.5+67.6
- FrontierMath Tiers 1-3 (v2)GPT-5.5 by 50.934.485.3+50.9
- HMMT Feb. 2026GPT-5.5 by 13.285.398.5+13.2
- LiveBench · MathematicsGPT-5.5 by 5.590.495.9+5.5
- AIME 2026GPT-5.5 by 4.995.1100+4.9
Following instructionsGPT-5.5
- IFBenchGPT-5.5 by 17.95875.9+17.9
- LiveBench · Instruction FollowingGPT-5.5 by 8.162.570.7+8.1
Other results23 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.
- CyberGymGPT-5.5 by 31.250.681.8+31.2
- frontiermath_tier_4_v1GPT-5.5 by 31.24.235.4+31.2
- Artificial Analysis Coding IndexGPT-5.5 by 27.147.874.9+27.1
- Terminal-Bench 2.0GPT-5.5 by 23.459.382.7+23.4
- AA Agentic IndexClaude Opus 4.5 by 22.359.637.3+22.3
- ARC-AGI-1GPT-5.5 by 138295+13
- ToolathlonGPT-5.5 by 12.143.555.6+12.1
- SimpleVQAClaude Opus 4.5 by 11.169.758.6+11.1
- AA IntelligenceGPT-5.5 by 9.329.138.4+9.3
- HLE (with tools)GPT-5.5 by 943.252.2+9
- NL2RepoGPT-5.5 by 7.543.250.7+7.5
- livebench_data_analysisGPT-5.5 by 7.274.481.6+7.2
- AA-OmniscienceGPT-5.5 by 6.51420.5+6.5
- livebench_languageGPT-5.5 by 6.181.387.4+6.1
- Finance AgentGPT-5.5 by 4.855.260+4.8
- LiveBenchGPT-5.5 by 4.77680.7+4.7
- τ²-Bench Telecom (AA run)GPT-5.5 by 4.489.593.9+4.4
- HMMT Nov. 2025GPT-5.5 by 3.293.396.5+3.2
- WorldVQAClaude Opus 4.5 by 2.236.834.6+2.2
- SuperGPQAGPT-5.5 by 2.170.672.7+2.1
- vectara_factual_consistencyGPT-5.5 by 1.689.190.7+1.6
- vectara_avg_summary_lengthGPT-5.5 by 15.1 rating points114.5129.6+15.1 rating
- vectara_answer_rateGPT-5.5 by 1.398.7100+1.3
Questions people ask
Which is better, Claude Opus 4.5 or GPT-5.5?
GPT-5.5 wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. Claude Opus 4.5 wins none, but costs 14% less.
Which is better for coding?
GPT-5.5. It wins 6 of the 6 coding tests both models report; Claude Opus 4.5 wins none.
Which is cheaper?
Claude Opus 4.5 costs $5.00 per million input tokens and $25.00 per million output tokens; GPT-5.5 costs $5.00 and $30.00. That makes Claude Opus 4.5 about 14% cheaper for the same work.
How do you compare the two?
We use the 52 benchmark tests both models have published scores on. The verdict counts the 29 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 23 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.