Claude Sonnet 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 Sonnet 4.5 is cheaper.
Scores updated · 44 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 own06GPT-5.56 of 6 tests
- CodingWriting and fixing software05GPT-5.55 of 5 tests
- ReasoningHard problems that need careful thinking05GPT-5.55 of 5 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
- MathCompetition and research-level math02GPT-5.52 of 2 tests
- Long documentsFinding answers in very long texts01GPT-5.51 of 1 test
- Following instructionsDoing exactly what it is asked01GPT-5.51 of 1 test
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 4.5 costs 49% 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
- Abstract visual puzzles that people can solveARC-AGI-2+71.4points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+70.1points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+61.4points ahead
Where Claude Sonnet 4.5 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+39.9points ahead
Every test, side by side
All 44 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.5
- Terminal-Bench 2.1GPT-5.5 by 28.555.884.3+28.5
- Terminal-Bench HardGPT-5.5 by 2535.660.6+25
- SWE-bench ProGPT-5.5 by 1543.658.6+15
- LMArena · WebDevGPT-5.5 by 119 rating points13941513+119 rating
- SciCodeGPT-5.5 by 10.145.755.8+10.1
AgentsGPT-5.5
- BrowseCompGPT-5.5 by 60.324.184.4+60.3
- MCP AtlasGPT-5.5 by 31.543.875.3+31.5
- GDPValGPT-5.5 by 22.220.542.7+22.2
- OSWorld-VerifiedGPT-5.5 by 17.361.478.7+17.3
- Terminal-Bench 4.0GPT-5.5 by 14.6014.6+14.6
- τ-Bench V3 · BankingGPT-5.5 by 14.524.539+14.5
ReasoningGPT-5.5
- ARC-AGI-2GPT-5.5 by 71.413.685+71.4
- Humanity's Last ExamGPT-5.5 by 2817.845.8+28
- CritPtGPT-5.5 by 261.127.1+26
- SimpleBenchGPT-5.5 by 14.754.369+14.7
- GPQA DiamondGPT-5.5 by 10.183.493.5+10.1
FactsGPT-5.5
- AA-Omniscience · Non-hallucinationClaude Sonnet 4.5 by 39.950.911+39.9
- SimpleQA VerifiedGPT-5.5 by 32.330.763+32.3
- AA-Omniscience · AccuracyGPT-5.5 by 25.132.958+25.1
- Vectara HHEM hallucination ratelower is betterGPT-5.5 by 2.7129.3+2.7
Images and chartsGPT-5.5
MathGPT-5.5
- FrontierMath Tier 4GPT-5.5 by 70.12.472.5+70.1
- FrontierMath Tiers 1-3 (v2)GPT-5.5 by 61.423.985.3+61.4
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.
- Terminal-Bench 2.0GPT-5.5 by 32.75082.7+32.7
- ARC-AGI-1GPT-5.5 by 31.363.795+31.3
- frontiermath_tier_4_v1GPT-5.5 by 31.24.235.4+31.2
- Artificial Analysis Coding IndexGPT-5.5 by 22.852.174.9+22.8
- AA-OmniscienceGPT-5.5 by 20.6-0.120.5+20.6
- HLE (with tools)GPT-5.5 by 20.23252.2+20.2
- AA Agentic IndexGPT-5.5 by 19.817.537.3+19.8
- matharena_visual_math_overallGPT-5.5 by 19.175.894.9+19.1
- AA IntelligenceGPT-5.5 by 17.720.738.4+17.7
- τ²-Bench Telecom (AA run)GPT-5.5 by 15.878.193.9+15.8
- HMMT Nov. 2025GPT-5.5 by 14.881.796.5+14.8
- ToolathlonGPT-5.5 by 14.64155.6+14.6
- HealthBenchGPT-5.5 by 12.344.256.5+12.3
- RealWorldQAGPT-5.5 by 11.970.382.2+11.9
- vectara_answer_rateGPT-5.5 by 4.495.6100+4.4
- vectara_factual_consistencyGPT-5.5 by 2.78890.7+2.7
- SimpleVQAGPT-5.5 by 157.658.6+1
- vectara_avg_summary_lengthtie127.8129.6tie
Questions people ask
Which is better, Claude Sonnet 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 Sonnet 4.5 wins none, but costs 49% less.
Which is better for coding?
GPT-5.5. It wins 5 of the 5 coding tests both models report; Claude Sonnet 4.5 wins none.
Which is cheaper?
Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens; GPT-5.5 costs $5.00 and $30.00. That makes Claude Sonnet 4.5 about 49% cheaper for the same work.
How do you compare the two?
We use the 44 benchmark tests both models have published scores on. The verdict counts the 26 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.