GPT-4.1 mini vs GPT-4o
Wins 2 of 7 areas
Reasoning · Images and charts
Wins 2 of 7 areas
Facts · Long documents
The two are evenly matched.GPT-4.1 mini is better at reasoning and images and charts; GPT-4o at facts and long documents.
Scores updated · 36 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.
- FactsGetting facts right instead of making them up03GPT-4o3 of 3 tests
- Long documentsFinding answers in very long texts01GPT-4o1 of 1 test
- ReasoningHard problems that need careful thinking20GPT-4.1 mini2 of 4 tests · 2 ties
- Images and chartsUnderstanding pictures, charts and video21GPT-4.1 mini2 of 4 tests · 1 tie
- CodingWriting and fixing software11Even1 each · 1 tie
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
- Following instructionsDoing exactly what it is asked11Even1 each
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 mini costs 90% 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-4o pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+54.8points ahead
- Short factual questions, answered correctlySimpleQA Verified+13.3points ahead
- Reasons across sets of long documentsAA-LCR+12points ahead
Where GPT-4.1 mini pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+13.8points ahead
- Math problems shown in pictures and chartsMathVista+11.7points ahead
- Code for real scientific research problemsSciCode+7.1points ahead
Every test, side by side
All 36 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- SWE-bench VerifiedGPT-4o by 9.623.633.2+9.6
- SciCodeGPT-4.1 mini by 7.140.433.3+7.1
- Terminal-Bench Hardtie7.68.3tie
ReasoningGPT-4.1 mini
- GPQA DiamondGPT-4.1 mini by 13.866.452.6+13.8
- Humanity's Last ExamGPT-4.1 mini by 3.251.8+3.2
- ARC-AGI-2tie00tie
- CritPttie00tie
FactsGPT-4o
- AA-Omniscience · Non-hallucinationGPT-4o by 54.87.362.1+54.8
- SimpleQA VerifiedGPT-4o by 13.312.726+13.3
- AA-Omniscience · AccuracyGPT-4o by 3.420.323.7+3.4
Images and chartsGPT-4.1 mini
- MathVistaGPT-4.1 mini by 11.773.161.4+11.7
- MMMU-ProGPT-4.1 mini by 2.458.756.3+2.4
- CharXiv (RQ)GPT-4o by 256.858.8+2
- MMMUtie72.772.2tie
Following instructionsEven
- Multi-ChallengeGPT-4o by 4.535.840.3+4.5
- IFBenchGPT-4.1 mini by 2.338.336+2.3
Other results18 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 43.1-53.6-10.5+43.1
- AIME 2025GPT-4.1 mini by 28.540.211.7+28.5
- τ²-Bench Telecom (AA run)GPT-4.1 mini by 2452.928.9+24
- TAU-bench (airline)GPT-4o by 6.83642.8+6.8
- AA Agentic IndexGPT-4o by 6.61.88.4+6.6
- TAU-bench (retail)GPT-4o by 4.555.860.3+4.5
- Aider-PolyglotGPT-4.1 mini by 434.730.7+4
- Artificial Analysis Coding IndexGPT-4o by 420.224.2+4
- IFEvalGPT-4.1 mini by 3.184.181+3.1
- bbqGPT-4o by 392.195.1+3
- AA IntelligenceGPT-4.1 mini by 2.910.27.3+2.9
- MMMLUGPT-4o by 2.978.581.4+2.9
- HarmBenchGPT-4.1 mini by 2.785.682.9+2.7
- AIR-Bench 2024GPT-4o by 260.462.4+2
- simple_safety_testsGPT-4.1 mini by 1.510098.5+1.5
- ARC-AGI-1GPT-4o by 13.54.5+1
- anthropic_red_teamtie99.399.1tie
- XSTesttie97.497.3tie
Questions people ask
Which is better, GPT-4.1 mini or GPT-4o?
GPT-4.1 mini and GPT-4o each win two of the seven areas where both have results. GPT-4.1 mini wins reasoning and images and charts; GPT-4o wins facts and long documents. GPT-4.1 mini costs 90% less. They are level on coding, agents and following instructions.
Which is better for coding?
Neither. They win 1 coding test each of the 3 both models report, and 1 is a tie.
Which is cheaper?
GPT-4.1 mini costs $0.40 per million input tokens and $1.60 per million output tokens; GPT-4o costs $5.00 and $15.00. That makes GPT-4.1 mini about 90% cheaper for the same work.
How do you compare the two?
We use the 36 benchmark tests both models have published scores on. The verdict counts the 18 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 18 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.