GPT-4o vs GPT-5 nano
Wins 2 of 7 areas
Facts · Long documents
Wins 5 of 7 areas
Coding · Agents · Reasoning · Images and charts · Following instructions
GPT-5 nano is the stronger all-rounder.GPT-4o is better at facts.
Scores updated · 31 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 thinking03GPT-5 nano3 of 4 tests · 1 tie
- CodingWriting and fixing software03GPT-5 nano3 of 3 tests
- AgentsCarrying out multi-step tasks on its own01GPT-5 nano1 of 1 test
- Images and chartsUnderstanding pictures, charts and video01GPT-5 nano1 of 1 test
- Following instructionsDoing exactly what it is asked01GPT-5 nano1 of 1 test
- FactsGetting facts right instead of making them up30GPT-4o3 of 4 tests · 1 tie
- Long documentsFinding answers in very long texts10GPT-4o1 of 1 test
Agents, images and charts, 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 nano costs 98% 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 nano pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+31.6points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+15points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+7.7points ahead
Where GPT-4o pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+21.1points ahead
- Short factual questions, answered correctlySimpleQA Verified+14.3points ahead
- Reasons across sets of long documentsAA-LCR+11points ahead
Every test, side by side
All 31 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5 nano
- Terminal-Bench HardGPT-5 nano by 3.88.312.1+3.8
- SciCodeGPT-5 nano by 3.333.336.6+3.3
- SWE-bench VerifiedGPT-5 nano by 1.633.234.8+1.6
ReasoningGPT-5 nano
- GPQA DiamondGPT-5 nano by 1552.667.6+15
- Humanity's Last ExamGPT-5 nano by 7.71.89.5+7.7
- ARC-AGI-2GPT-5 nano by 2.602.6+2.6
- CritPttie00tie
FactsGPT-4o
- AA-Omniscience · Non-hallucinationGPT-4o by 21.162.141+21.1
- SimpleQA VerifiedGPT-4o by 14.32611.7+14.3
- AA-Omniscience · AccuracyGPT-4o by 4.623.719.1+4.6
- Vectara HHEM hallucination ratelower is bettertie9.610.5tie
Following instructionsGPT-5 nano
- IFBenchGPT-5 nano by 31.63667.6+31.6
Other results16 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 2025GPT-5 nano by 73.511.785.2+73.5
- AIR-Bench 2024GPT-5 nano by 25.462.487.8+25.4
- AA-OmniscienceGPT-4o by 18.2-10.5-28.7+18.2
- HarmBenchGPT-5 nano by 15.582.998.4+15.5
- AA Agentic IndexGPT-5 nano by 14.88.423.2+14.8
- ARC-AGI-1GPT-5 nano by 12.24.516.7+12.2
- τ²-Bench Telecom (AA run)GPT-5 nano by 7.628.936.5+7.6
- vectara_answer_rateGPT-5 nano by 6.293.8100+6.2
- AA IntelligenceGPT-5 nano by 5.77.313+5.7
- Artificial Analysis Coding IndexGPT-4o by 3.924.220.3+3.9
- bbqGPT-5 nano by 2.595.197.6+2.5
- vectara_avg_summary_lengthGPT-5 nano by 19.1 rating points86.6105.7+19.1 rating
- simple_safety_testsGPT-5 nano by 1.598.5100+1.5
- vectara_factual_consistencytie90.489.5tie
- anthropic_red_teamtie99.199.5tie
- XSTesttie97.397.6tie
Questions people ask
Which is better, GPT-4o or GPT-5 nano?
GPT-5 nano wins five of the seven areas where both have results: coding, agents, reasoning, images and charts and following instructions. GPT-4o wins facts and long documents.
Which is better for coding?
GPT-5 nano. 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; GPT-5 nano costs $0.05 and $0.40. That makes GPT-5 nano about 98% cheaper for the same work.
How do you compare the two?
We use the 31 benchmark tests both models have published scores on. The verdict counts the 15 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 16 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.