GPT-5.5 vs Claude Sonnet 5.5
Wins 3 of 8 areas
Facts · Long documents · Following instructions
Wins 4 of 8 areas
Coding · Agents · Reasoning · Math
Claude Sonnet 5.5 wins more areas, narrowly.GPT-5.5 is better at facts.
Both rank among the ten best models we track in math.
Scores updated · 33 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 software05Claude Sonnet 5.55 of 5 tests
- ReasoningHard problems that need careful thinking04Claude Sonnet 5.54 of 4 tests
- MathCompetition and research-level math02Claude Sonnet 5.52 of 3 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own02Claude Sonnet 5.52 of 2 tests
- FactsGetting facts right instead of making them up21GPT-5.52 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
- Images and chartsUnderstanding pictures, charts and video00Even0 each · 1 tie
Images and charts, 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 5.5 costs 66% 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 Sonnet 5.5 pulls ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+49points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+42points ahead
- Real work tasks from 44 professionsGDPVal+24.4points ahead
Where GPT-5.5 pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+16.5points ahead
- Follows detailed instructions exactlyLiveBench · Instruction Following+13.9points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+4points ahead
Every test, side by side
All 33 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude Sonnet 5.5
- LMArena · WebDevClaude Sonnet 5.5 by 281 rating points14871768+281 rating
- SWE-bench ProClaude Sonnet 5.5 by 22.758.681.3+22.7
- LiveBench · CodingClaude Sonnet 5.5 by 9.282.291.4+9.2
- SciCodeClaude Sonnet 5.5 by 5.255.861+5.2
- LiveBench · Agentic CodingClaude Sonnet 5.5 by 2.35456.3+2.3
AgentsClaude Sonnet 5.5
- Terminal-Bench 4.0Claude Sonnet 5.5 by 4914.663.6+49
- GDPValClaude Sonnet 5.5 by 24.442.566.9+24.4
ReasoningClaude Sonnet 5.5
- Humanity's Last ExamClaude Sonnet 5.5 by 9.245.855+9.2
- SimpleBenchClaude Sonnet 5.5 by 6.96975.9+6.9
- CritPtClaude Sonnet 5.5 by 4.327.131.4+4.3
- LiveBench · ReasoningClaude Sonnet 5.5 by 1.989.791.6+1.9
FactsGPT-5.5
- AA-Omniscience · Non-hallucinationClaude Sonnet 5.5 by 421153+42
- SimpleQA VerifiedGPT-5.5 by 16.56346.5+16.5
- AA-Omniscience · AccuracyGPT-5.5 by 45854+4
MathClaude Sonnet 5.5
- FrontierMath Tier 4Claude Sonnet 5.5 by 872.580.5+8
- FrontierMath Tiers 1-3 (v2)Claude Sonnet 5.5 by 3.585.388.8+3.5
- LiveBench · Mathematicstie95.996.1tie
Following instructionsGPT-5.5
- LiveBench · Instruction FollowingGPT-5.5 by 13.970.756.8+13.9
Other results13 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.
- livebench_data_analysisGPT-5.5 by 22.181.659.5+22.1
- AA IntelligenceClaude Sonnet 5.5 by 17.638.456+17.6
- HealthBench ProfessionalClaude Sonnet 5.5 by 17.451.869.2+17.4
- HealthBench Professional length-adjustedClaude Sonnet 5.5 by 17.451.869.2+17.4
- HLE (with tools)Claude Sonnet 5.5 by 12.352.264.5+12.3
- AA-OmniscienceClaude Sonnet 5.5 by 11.820.532.3+11.8
- OfficeQA ProClaude Sonnet 5.5 by 11.554.165.6+11.5
- Legal Agent BenchmarkClaude Sonnet 5.5 by 9.62.111.7+9.6
- livebench_languageGPT-5.5 by 9.487.478+9.4
- HealthBenchClaude Sonnet 5.5 by 8.956.565.4+8.9
- Program BenchClaude Sonnet 5.5 by 8.970.879.7+8.9
- Toolathlon VerifiedClaude Sonnet 5.5 by 4.373.577.8+4.3
- DeepSWE 1.1Claude Sonnet 5.5 by 46771+4
Questions people ask
Which is better, GPT-5.5 or Claude Sonnet 5.5?
Claude Sonnet 5.5 wins four of the eight areas we test: coding, agents, reasoning and math. GPT-5.5 wins facts, long documents and following instructions. They are level on images and charts.
Which is better for coding?
Claude Sonnet 5.5. It wins 5 of the 5 coding tests both models report; GPT-5.5 wins none.
Which is cheaper?
GPT-5.5 costs $5.00 per million input tokens and $30.00 per million output tokens; Claude Sonnet 5.5 costs $2.00 and $10.00. That makes Claude Sonnet 5.5 about 66% cheaper for the same work.
How do you compare the two?
We use the 33 benchmark tests both models have published scores on. The verdict counts the 20 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 13 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.