Claude 3.7 Sonnet vs GPT-4o
Wins 5 of 7 areas
Coding · Agents · Reasoning · Images and charts · Following instructions
Wins 1 of 7 areas
Long documents
Claude 3.7 Sonnet is the stronger all-rounder.GPT-4o is better at long documents.
Scores updated · 36 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.
- CodingWriting and fixing software30Claude 3.7 Sonnet3 of 3 tests
- ReasoningHard problems that need careful thinking20Claude 3.7 Sonnet2 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video20Claude 3.7 Sonnet2 of 2 tests
- Following instructionsDoing exactly what it is asked20Claude 3.7 Sonnet2 of 2 tests
- AgentsCarrying out multi-step tasks on its own10Claude 3.7 Sonnet1 of 1 test
- Long documentsFinding answers in very long texts01GPT-4o1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Agents and long documents 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 3.7 Sonnet costs 10% 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 Claude 3.7 Sonnet pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+37.1points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+24.6points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+12.9points ahead
Where GPT-4o pulls ahead
- Reasons across sets of long documentsAA-LCR+4.3points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+2.1points ahead
Every test, side by side
All 36 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingClaude 3.7 Sonnet
- SWE-bench VerifiedClaude 3.7 Sonnet by 37.170.333.2+37.1
- Terminal-Bench HardClaude 3.7 Sonnet by 12.921.28.3+12.9
- SciCodeClaude 3.7 Sonnet by 740.333.3+7
ReasoningClaude 3.7 Sonnet
- GPQA DiamondClaude 3.7 Sonnet by 24.677.252.6+24.6
- Humanity's Last ExamClaude 3.7 Sonnet by 7.99.71.8+7.9
- CritPttie0.90tie
FactsEven
- AA-Omniscience · AccuracyClaude 3.7 Sonnet by 4.32823.7+4.3
- AA-Omniscience · Non-hallucinationGPT-4o by 2.16062.1+2.1
Images and chartsClaude 3.7 Sonnet
Following instructionsClaude 3.7 Sonnet
- IFBenchClaude 3.7 Sonnet by 12.348.336+12.3
- Multi-ChallengeClaude 3.7 Sonnet by 11.351.640.3+11.3
Other results22 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.
- AIME 2025Claude 3.7 Sonnet by 43.154.811.7+43.1
- Aider-PolyglotClaude 3.7 Sonnet by 29.760.430.7+29.7
- metr_hcastClaude 3.7 Sonnet by 29.254.225+29.2
- AA Agentic IndexClaude 3.7 Sonnet by 28.6378.4+28.6
- τ²-Bench Telecom (AA run)Claude 3.7 Sonnet by 25.854.728.9+25.8
- ARC-AGI-1Claude 3.7 Sonnet by 24.128.64.5+24.1
- MATH-500 (EM)Claude 3.7 Sonnet by 21.696.274.6+21.6
- TAU-bench (retail)Claude 3.7 Sonnet by 20.981.260.3+20.9
- AIR-Bench 2024Claude 3.7 Sonnet by 19.481.862.4+19.4
- TAU-bench (airline)Claude 3.7 Sonnet by 15.658.442.8+15.6
- Artificial Analysis Coding IndexClaude 3.7 Sonnet by 12.236.424.2+12.2
- IFEvalClaude 3.7 Sonnet by 12.293.281+12.2
- AA IntelligenceClaude 3.7 Sonnet by 10.417.77.3+10.4
- AA-OmniscienceClaude 3.7 Sonnet by 9.8-0.7-10.5+9.8
- MMMLUClaude 3.7 Sonnet by 4.786.181.4+4.7
- bbqGPT-4o by 392.195.1+3
- simple_safety_testsClaude 3.7 Sonnet by 1.510098.5+1.5
- HarmBenchClaude 3.7 Sonnet by 1.484.382.9+1.4
- XSTesttie96.497.3tie
- anthropic_red_teamtie99.799.1tie
- metr_swaatie98.799.2tie
- metr_re_benchtie00tie
Questions people ask
Which is better, Claude 3.7 Sonnet or GPT-4o?
Claude 3.7 Sonnet wins five of the seven areas where both have results: coding, agents, reasoning, images and charts and following instructions. GPT-4o wins long documents. They are level on facts.
Which is better for coding?
Claude 3.7 Sonnet. It wins 3 of the 3 coding tests both models report; GPT-4o wins none.
Which is cheaper?
Claude 3.7 Sonnet costs $3.00 per million input tokens and $15.00 per million output tokens; GPT-4o costs $5.00 and $15.00. That makes Claude 3.7 Sonnet about 10% cheaper for the same work.
How do you compare the two?
We use the 36 benchmark tests both models have published scores on. The verdict counts the 14 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 22 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.