GPT-4.1 mini vs GPT-4.1 nano
Wins 7 of 7 areas
Coding · Agents · Reasoning · Facts · Images and charts · Long documents · Following instructions
Wins 0 of 7 areas
—
GPT-4.1 mini is the stronger all-rounder.GPT-4.1 nano is cheaper.
Scores updated · 35 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 video40GPT-4.1 mini4 of 4 tests
- CodingWriting and fixing software30GPT-4.1 mini3 of 3 tests
- ReasoningHard problems that need careful thinking20GPT-4.1 mini2 of 4 tests · 2 ties
- Following instructionsDoing exactly what it is asked20GPT-4.1 mini2 of 2 tests
- FactsGetting facts right instead of making them up21GPT-4.1 mini2 of 3 tests
- AgentsCarrying out multi-step tasks on its own10GPT-4.1 mini1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts10GPT-4.1 mini1 of 1 test
Long documents rests on a single test.
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 75% 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-4.1 mini pulls ahead
- Reasons across sets of long documentsAA-LCR+23.7points ahead
- Keeps track of context across a multi-turn chatMulti-Challenge+20.8points ahead
- Harder college exam questions with imagesMMMU-Pro+18.6points ahead
Where GPT-4.1 nano pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+10.1points ahead
Every test, side by side
All 35 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-4.1 mini
- SciCodeGPT-4.1 mini by 14.540.425.9+14.5
- Terminal-Bench 2.1GPT-4.1 mini by 6.410.13.7+6.4
- Terminal-Bench HardGPT-4.1 mini by 3.87.63.8+3.8
AgentsGPT-4.1 mini
- τ-Bench V3 · BankingGPT-4.1 mini by 1.95.43.5+1.9
- GDPValtie00tie
ReasoningGPT-4.1 mini
- GPQA DiamondGPT-4.1 mini by 15.266.451.2+15.2
- Humanity's Last ExamGPT-4.1 mini by 1.253.8+1.2
- ARC-AGI-2tie00tie
- CritPttie00tie
FactsGPT-4.1 mini
- AA-Omniscience · Non-hallucinationGPT-4.1 nano by 10.17.317.4+10.1
- SimpleQA VerifiedGPT-4.1 mini by 6.712.76+6.7
- AA-Omniscience · AccuracyGPT-4.1 mini by 6.620.313.7+6.6
Images and chartsGPT-4.1 mini
- MMMU-ProGPT-4.1 mini by 18.658.740.1+18.6
- MMMUGPT-4.1 mini by 17.372.755.4+17.3
- MathVistaGPT-4.1 mini by 16.973.156.2+16.9
- CharXiv (RQ)GPT-4.1 mini by 16.356.840.5+16.3
Following instructionsGPT-4.1 mini
- Multi-ChallengeGPT-4.1 mini by 20.835.815+20.8
- IFBenchGPT-4.1 mini by 6.338.332+6.3
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.
- τ²-Bench Telecom (AA run)GPT-4.1 mini by 35.652.917.3+35.6
- TAU-bench (retail)GPT-4.1 mini by 33.255.822.6+33.2
- Aider-PolyglotGPT-4.1 mini by 24.934.79.8+24.9
- TAU-bench (airline)GPT-4.1 mini by 223614+22
- MMMLUGPT-4.1 mini by 11.678.566.9+11.6
- IFEvalGPT-4.1 mini by 9.684.174.5+9.6
- Artificial Analysis Coding IndexGPT-4.1 mini by 9.120.211.1+9.1
- bbqGPT-4.1 mini by 4.692.187.5+4.6
- AA-OmniscienceGPT-4.1 mini by 4-53.6-57.6+4
- ARC-AGI-1GPT-4.1 mini by 3.53.50+3.5
- AA IntelligenceGPT-4.1 mini by 2.410.27.8+2.4
- XSTestGPT-4.1 mini by 1.497.496+1.4
- HarmBenchGPT-4.1 nano by 1.285.686.8+1.2
- AIR-Bench 2024GPT-4.1 nano by 1.160.461.5+1.1
- simple_safety_testsGPT-4.1 mini by 110099+1
- anthropic_red_teamtie99.399.6tie
Questions people ask
Which is better, GPT-4.1 mini or GPT-4.1 nano?
GPT-4.1 mini wins all seven areas where both have results: coding, agents, reasoning, facts, images and charts, long documents and following instructions. GPT-4.1 nano wins none, but costs 75% less.
Which is better for coding?
GPT-4.1 mini. It wins 3 of the 3 coding tests both models report; GPT-4.1 nano wins none.
Which is cheaper?
GPT-4.1 mini costs $0.40 per million input tokens and $1.60 per million output tokens; GPT-4.1 nano costs $0.10 and $0.40. That makes GPT-4.1 nano about 75% cheaper for the same work.
How do you compare the two?
We use the 35 benchmark tests both models have published scores on. The verdict counts the 19 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.