GPT-4o vs o1
Wins 0 of 7 areas
—
Wins 7 of 7 areas
Coding · Agents · Reasoning · Facts · Images and charts · Long documents · Following instructions
o1 is the stronger all-rounder.GPT-4o is cheaper.
Scores updated · 46 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 software03o13 of 3 tests
- ReasoningHard problems that need careful thinking02o12 of 3 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video02o12 of 2 tests
- FactsGetting facts right instead of making them up12o12 of 3 tests
- AgentsCarrying out multi-step tasks on its own01o11 of 1 test
- Long documentsFinding answers in very long texts01o11 of 1 test
- Following instructionsDoing exactly what it is asked01o11 of 1 test
Agents, 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-4o costs 73% 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 o1 pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+34.3points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+22.1points ahead
- Short factual questions, answered correctlySimpleQA Verified+15.1points ahead
Where GPT-4o pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+31.7points ahead
Every test, side by side
All 46 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Codingo1
- SWE-bench Verifiedo1 by 8.133.241.3+8.1
- Terminal-Bench Hardo1 by 4.68.312.9+4.6
- SciCodeo1 by 2.533.335.8+2.5
Reasoningo1
- GPQA Diamondo1 by 22.152.674.7+22.1
- Humanity's Last Examo1 by 5.21.87+5.2
- CritPttie00.3tie
Factso1
- AA-Omniscience · Non-hallucinationGPT-4o by 31.762.130.4+31.7
- SimpleQA Verifiedo1 by 15.12641.1+15.1
- AA-Omniscience · Accuracyo1 by 10.823.734.5+10.8
Images and chartso1
Other results32 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.
- Codeforceso1 by 1302 rating points7592061+1302 rating
- Codeforces (Rating)o1 by 1302 rating points7592061+1302 rating
- Codeforces (Percentile)o1 by 7323.696.6+73
- AIME 2025o1 by 7011.781.7+70
- AIME 2024o1 by 69.99.379.2+69.9
- τ²-Bench Telecom (AA run)o1 by 33.728.962.6+33.7
- Aider-Polygloto1 by 3130.761.7+31
- LiveCodeBench (Pass@1-COT)o1 by 29.234.263.4+29.2
- AA Agentic Indexo1 by 22.78.431.1+22.7
- MATH-500 (EM)o1 by 21.874.696.4+21.8
- MMLUo1 by 1873.891.8+18
- AIR-Bench 2024o1 by 17.662.480+17.6
- metr_hcasto1 by 17.62542.6+17.6
- Artificial Analysis Coding Indexo1 by 15.524.239.7+15.5
- HarmBencho1 by 13.482.996.3+13.4
- MATHo1 by 11.185.396.4+11.1
- TAU-bench (retail)o1 by 10.560.370.8+10.5
- AA Intelligenceo1 by 7.97.315.2+7.9
- TAU-bench (airline)o1 by 7.242.850+7.2
- DROP (3-shot F1)o1 by 6.583.790.2+6.5
- MMMLUo1 by 6.381.487.7+6.3
- GSM8Ko1 by 6.290.997.1+6.2
- SimpleQAo1 by 4.238.242.4+4.2
- bbqo1 by 2.295.197.3+2.2
- HumanEvalGPT-4o by 2.190.288.1+2.1
- anthropic_red_teamtie99.198.3tie
- metr_swaatie99.2100tie
- AA-Omnisciencetie-10.5-11.1tie
- simple_safety_teststie98.599tie
- MGSMtie90.590.8tie
- XSTesttie97.397tie
- metr_re_benchtie00tie
Questions people ask
Which is better, GPT-4o or o1?
o1 wins all seven areas where both have results: coding, agents, reasoning, facts, images and charts, long documents and following instructions. GPT-4o wins none, but costs 73% less.
Which is better for coding?
o1. It wins 3 of the 3 coding tests both models report; GPT-4o wins none.
Which is cheaper?
GPT-4o costs $5.00 per million input tokens and $15.00 per million output tokens; o1 costs $15.00 and $60.00. That makes GPT-4o about 73% cheaper for the same work.
How do you compare the two?
We use the 46 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 32 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.