GPT-5.5 vs Claude Opus 5
Wins 2 of 8 areas
Long documents · Following instructions
Wins 5 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts
Claude Opus 5 is the stronger all-rounder.GPT-5.5 is better at long documents.
Both rank among the ten best models we track in reasoning, coding, agents and math.
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.
- ReasoningHard problems that need careful thinking06Claude Opus 56 of 7 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own06Claude Opus 56 of 6 tests
- CodingWriting and fixing software04Claude Opus 54 of 6 tests · 2 ties
- Images and chartsUnderstanding pictures, charts and video02Claude Opus 52 of 2 tests
- FactsGetting facts right instead of making them up12Claude Opus 52 of 3 tests
- Long documentsFinding answers in very long texts10GPT-5.51 of 1 test
- Following instructionsDoing exactly what it is asked10GPT-5.51 of 1 test
- MathCompetition and research-level math00Even0 each · 3 ties
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 Opus 5 costs 14% 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 5 pulls ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+34.4points ahead
- Solves new puzzle games it has never seenARC-AGI-3+29.8points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+28.2points ahead
Where GPT-5.5 pulls ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+6.9points ahead
- Reasons across sets of long documentsAA-LCR+5points ahead
- Short factual questions, answered correctlySimpleQA Verified+3.1points 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.
CodingClaude Opus 5
- SWE-bench ProClaude Opus 5 by 20.658.679.2+20.6
- LMArena · WebDevClaude Opus 5 by 179 rating points15131692+179 rating
- LiveBench · Agentic CodingClaude Opus 5 by 11.25465.2+11.2
- Terminal-Bench 2.1Claude Opus 5 by 4.884.389.1+4.8
- LiveBench · Codingtie82.281.5tie
- SciCodetie55.856.4tie
AgentsClaude Opus 5
- Terminal-Bench 4.0Claude Opus 5 by 34.414.649+34.4
- GDPValClaude Opus 5 by 18.542.761.2+18.5
- MCP AtlasClaude Opus 5 by 10.575.385.8+10.5
- BrowseCompClaude Opus 5 by 6.484.490.8+6.4
- OSWorld-VerifiedClaude Opus 5 by 4.778.783.4+4.7
- τ-Bench V3 · BankingClaude Opus 5 by 3.13942.1+3.1
ReasoningClaude Opus 5
- ARC-AGI-3Claude Opus 5 by 29.80.430.2+29.8
- SimpleBenchClaude Opus 5 by 11.66980.6+11.6
- Humanity's Last ExamClaude Opus 5 by 9.145.854.9+9.1
- ARC-AGI-2Claude Opus 5 by 5.48590.4+5.4
- CritPtClaude Opus 5 by 227.129.1+2
- LiveBench · ReasoningClaude Opus 5 by 1.589.791.2+1.5
- GPQA Diamondtie93.593.2tie
FactsClaude Opus 5
- AA-Omniscience · Non-hallucinationClaude Opus 5 by 28.21139.2+28.2
- SimpleQA VerifiedGPT-5.5 by 3.16359.9+3.1
- AA-Omniscience · AccuracyClaude Opus 5 by 2.95860.9+2.9
Images and chartsClaude Opus 5
- MMMU-ProClaude Opus 5 by 4.879.984.7+4.8
- LMArena · VisionClaude Opus 5 by 29 rating points12921321+29 rating
MathEven
- FrontierMath Tier 4tie72.573.2tie
- FrontierMath Tiers 1-3 (v2)tie85.385.6tie
- LiveBench · Mathematicstie95.995.7tie
Following instructionsGPT-5.5
- LiveBench · Instruction FollowingGPT-5.5 by 6.970.763.8+6.9
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.
- GDP (Surge AI)Claude Opus 5 by 60.624.985.5+60.6
- GDPval-AA v2 EloClaude Opus 5 by 333 rating points14911824+333 rating
- GDPval-AA v2Claude Opus 5 by 315 rating points15091824+315 rating
- JobBenchClaude Opus 5 by 27.438.365.7+27.4
- AA Agentic IndexClaude Opus 5 by 18.937.356.2+18.9
- HiL-Bench (Tools-allowed)Claude Opus 5 by 17.339.757+17.3
- AA-OmniscienceClaude Opus 5 by 16.620.537.1+16.6
- Program BenchClaude Opus 5 by 14.670.885.4+14.6
- OfficeQA ProClaude Opus 5 by 12.854.166.9+12.8
- AA IntelligenceClaude Opus 5 by 12.438.450.8+12.4
- HLE (with tools)Claude Opus 5 by 11.452.263.6+11.4
- Legal Agent BenchmarkClaude Opus 5 by 9.62.111.7+9.6
- HealthBench ProfessionalClaude Opus 5 by 851.859.8+8
- HealthBench Professional length-adjustedClaude Opus 5 by 851.859.8+8
- Toolathlon VerifiedClaude Opus 5 by 7.173.580.6+7.1
- livebench_data_analysisGPT-5.5 by 781.674.5+7
- GDPval-AA (Elo)GPT-5.5 by 61 rating points17691708+61 rating
- Agents' Last ExamClaude Opus 5 by 526.631.6+5
- Artificial Analysis Coding IndexClaude Opus 5 by 3.174.978+3.1
- ARC-AGI-1Claude Opus 5 by 2.59597.5+2.5
- DeepSWE 1.1Claude Opus 5 by 1.86768.8+1.8
- HealthBench length-adjustedClaude Opus 5 by 1.356.557.8+1.3
- livebench_languageClaude Opus 5 by 1.387.488.7+1.3
Questions people ask
Which is better, GPT-5.5 or Claude Opus 5?
Claude Opus 5 wins five of the eight areas we test: coding, agents, reasoning, facts and images and charts. GPT-5.5 wins long documents and following instructions. They are level on math.
Which is better for coding?
Claude Opus 5. It wins 4 of the 6 coding tests both models report; GPT-5.5 wins none, and 2 are ties.
Which is cheaper?
GPT-5.5 costs $5.00 per million input tokens and $30.00 per million output tokens; Claude Opus 5 costs $5.00 and $25.00. That makes Claude Opus 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.