GPT-5 Pro is the stronger all-rounder.Claude Sonnet 4.5 is cheaper.
Scores updated · 6 tests both models report · How we compare
Where each one wins
Tests won in each of the two 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.
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 87% 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 Pro pulls ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+31.9points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+17.1points ahead
- Common-sense trick questionsSimpleBench+7.3points ahead
Where Claude Sonnet 4.5 pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 6 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningGPT-5 Pro
- SimpleBenchGPT-5 Pro by 7.354.361.6+7.3
- ARC-AGI-2GPT-5 Pro by 4.713.618.3+4.7
MathGPT-5 Pro
- FrontierMath Tiers 1-3 (v2)GPT-5 Pro by 31.923.955.8+31.9
- FrontierMath Tier 4GPT-5 Pro by 17.12.419.5+17.1
Other results2 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.
- frontiermath_tier_4_v1GPT-5 Pro by 10.44.214.6+10.4
- ARC-AGI-1GPT-5 Pro by 6.563.770.2+6.5
Questions people ask
Which is better, Claude Sonnet 4.5 or GPT-5 Pro?
GPT-5 Pro wins all two areas where both have results: reasoning and math. Claude Sonnet 4.5 wins none, but costs 87% less.
Which is cheaper?
Claude Sonnet 4.5 costs $3.00 per million input tokens and $15.00 per million output tokens; GPT-5 Pro costs $15.00 and $120.00. That makes Claude Sonnet 4.5 about 87% cheaper for the same work.
How do you compare the two?
We use the 6 benchmark tests both models have published scores on. The verdict counts the 4 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 2 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.