Claude Sonnet 4 vs GPT-5.2
Wins 0 of 7 areas
—
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
GPT-5.2 is the stronger all-rounder.
Scores updated · 39 tests both models report · How we compare
Where each one wins
Tests won in each of the seven areas where both have results. 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 thinking04GPT-5.24 of 5 tests · 1 tie
- CodingWriting and fixing software14GPT-5.24 of 5 tests
- Images and chartsUnderstanding pictures, charts and video02GPT-5.22 of 2 tests
- AgentsCarrying out multi-step tasks on its own12GPT-5.22 of 3 tests
- Long documentsFinding answers in very long texts01GPT-5.21 of 1 test
- Following instructionsDoing exactly what it is asked01GPT-5.21 of 1 test
- FactsGetting facts right instead of making them up11Even1 each · 1 tie
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.
GPT-5.2 costs 13% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GPT-5.2 pulls ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+53.6points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+47points ahead
- Real work tasks from 44 professionsGDPVal+38.6points ahead
Where Claude Sonnet 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+52.1points ahead
- Long, multi-file coding tasks in real codebasesSWE-bench Pro+12.8points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+5.6points ahead
Every test, side by side
All 39 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.2
- Terminal-Bench HardGPT-5.2 by 15.931.147+15.9
- SWE-bench ProClaude Sonnet 4 by 12.842.729.9+12.8
- SciCodeGPT-5.2 by 12.14052.1+12.1
- SWE-bench MultilingualGPT-5.2 by 9.856.966.7+9.8
- SWE-bench VerifiedGPT-5.2 by 7.372.780+7.3
AgentsGPT-5.2
- BrowseCompGPT-5.2 by 53.612.265.8+53.6
- GDPValGPT-5.2 by 38.69.748.3+38.6
- τ-Bench V3 · BankingClaude Sonnet 4 by 5.616.711.1+5.6
ReasoningGPT-5.2
- ARC-AGI-2GPT-5.2 by 475.952.9+47
- Humanity's Last ExamGPT-5.2 by 2710.737.7+27
- GPQA DiamondGPT-5.2 by 12.677.790.3+12.6
- CritPtGPT-5.2 by 11.30.311.6+11.3
- SimpleBenchtie45.545.8tie
FactsEven
- AA-Omniscience · Non-hallucinationClaude Sonnet 4 by 52.170.918.8+52.1
- AA-Omniscience · AccuracyGPT-5.2 by 21.622.744.3+21.6
- Vectara HHEM hallucination ratelower is bettertie10.310.8tie
Images and chartsGPT-5.2
- MMMU-ProGPT-5.2 by 12.861.874.6+12.8
- LMArena · VisionGPT-5.2 by 55 rating points11911246+55 rating
Other results19 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.
- HMMT 2025GPT-5.2 by 83.515.999.4+83.5
- ARC-AGI-1GPT-5.2 by 46.24086.2+46.2
- AA Agentic IndexGPT-5.2 by 42.617.660.2+42.6
- τ²-BenchGPT-5.2 by 40.345.285.5+40.3
- AIME 2025GPT-5.2 by 29.570.5100+29.5
- HLE (with tools)GPT-5.2 by 25.220.345.5+25.2
- LiveCodeBenchGPT-5.2 by 20.568.589+20.5
- τ²-Bench Telecom (AA run)GPT-5.2 by 20.264.684.8+20.2
- frontiermath_tier_4_v1GPT-5.2 by 18.8018.8+18.8
- AA IntelligenceGPT-5.2 by 11.518.930.4+11.5
- Artificial Analysis Coding IndexGPT-5.2 by 11.137.648.7+11.1
- τ²-Bench (Retail)GPT-5.2 by 77582+7
- vectara_avg_summary_lengthGPT-5.2 by 40.5 rating points145.8186.3+40.5 rating
- MMMLUGPT-5.2 by 3.186.589.6+3.1
- MMLU-ProGPT-5.2 by 383.786.7+3
- vectara_answer_rateGPT-5.2 by 1.498.6100+1.4
- AA-OmniscienceClaude Sonnet 4 by 1.10.2-0.9+1.1
- vectara_factual_consistencytie89.789.2tie
- LiveBenchtie74.874.8tie
Questions people ask
Which is better, Claude Sonnet 4 or GPT-5.2?
GPT-5.2 wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Claude Sonnet 4 wins none. They are level on facts.
Which is better for coding?
GPT-5.2. It wins 4 of the 5 coding tests both models report; Claude Sonnet 4 wins 1.
Which is cheaper?
Claude Sonnet 4 costs $3.00 per million input tokens and $15.00 per million output tokens; GPT-5.2 costs $1.75 and $14.00. That makes GPT-5.2 about 13% cheaper for the same work.
How do you compare the two?
We use the 39 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 19 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.