GPT-4.1 nano vs GPT-4o
Wins 0 of 7 areas
—
Wins 5 of 7 areas
Coding · Facts · Images and charts · Long documents · Following instructions
GPT-4o is the stronger all-rounder.GPT-4.1 nano is cheaper.
Scores updated · 33 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.
- Images and chartsUnderstanding pictures, charts and video04GPT-4o4 of 4 tests
- FactsGetting facts right instead of making them up03GPT-4o3 of 3 tests
- CodingWriting and fixing software02GPT-4o2 of 2 tests
- Following instructionsDoing exactly what it is asked02GPT-4o2 of 2 tests
- Long documentsFinding answers in very long texts01GPT-4o1 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
- ReasoningHard problems that need careful thinking11Even1 each · 2 ties
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.
GPT-4.1 nano costs 98% 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-4o pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+44.7points ahead
- Reasons across sets of long documentsAA-LCR+35.7points ahead
- Keeps track of context across a multi-turn chatMulti-Challenge+25.3points ahead
Where GPT-4.1 nano pulls ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+2points ahead
Every test, side by side
All 33 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-4o
- SciCodeGPT-4o by 7.425.933.3+7.4
- Terminal-Bench HardGPT-4o by 4.53.88.3+4.5
ReasoningEven
- Humanity's Last ExamGPT-4.1 nano by 23.81.8+2
- GPQA DiamondGPT-4o by 1.451.252.6+1.4
- ARC-AGI-2tie00tie
- CritPttie00tie
FactsGPT-4o
- AA-Omniscience · Non-hallucinationGPT-4o by 44.717.462.1+44.7
- SimpleQA VerifiedGPT-4o by 20626+20
- AA-Omniscience · AccuracyGPT-4o by 1013.723.7+10
Images and chartsGPT-4o
- CharXiv (RQ)GPT-4o by 18.340.558.8+18.3
- MMMUGPT-4o by 16.855.472.2+16.8
- MMMU-ProGPT-4o by 16.240.156.3+16.2
- MathVistaGPT-4o by 5.256.261.4+5.2
Following instructionsGPT-4o
- Multi-ChallengeGPT-4o by 25.31540.3+25.3
- IFBenchGPT-4o by 43236+4
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.
- AA-OmniscienceGPT-4o by 47.1-57.6-10.5+47.1
- TAU-bench (retail)GPT-4o by 37.722.660.3+37.7
- TAU-bench (airline)GPT-4o by 28.81442.8+28.8
- Aider-PolyglotGPT-4o by 20.99.830.7+20.9
- MMMLUGPT-4o by 14.566.981.4+14.5
- Artificial Analysis Coding IndexGPT-4o by 13.111.124.2+13.1
- τ²-Bench Telecom (AA run)GPT-4o by 11.617.328.9+11.6
- bbqGPT-4o by 7.687.595.1+7.6
- IFEvalGPT-4o by 6.574.581+6.5
- ARC-AGI-1GPT-4o by 4.504.5+4.5
- HarmBenchGPT-4.1 nano by 3.986.882.9+3.9
- XSTestGPT-4o by 1.39697.3+1.3
- AIR-Bench 2024tie61.562.4tie
- AA Intelligencetie7.87.3tie
- anthropic_red_teamtie99.699.1tie
- simple_safety_teststie9998.5tie
Questions people ask
Which is better, GPT-4.1 nano or GPT-4o?
GPT-4o wins five of the seven areas where both have results: coding, facts, images and charts, long documents and following instructions. GPT-4.1 nano wins none, but costs 98% less. They are level on agents and reasoning.
Which is better for coding?
GPT-4o. It wins 2 of the 2 coding tests both models report; GPT-4.1 nano wins none.
Which is cheaper?
GPT-4.1 nano costs $0.10 per million input tokens and $0.40 per million output tokens; GPT-4o costs $5.00 and $15.00. That makes GPT-4.1 nano about 98% cheaper for the same work.
How do you compare the two?
We use the 33 benchmark tests both models have published scores on. The verdict counts the 17 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.