GPT-4.1 mini vs GPT-5
Wins 0 of 8 areas
—
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
GPT-5 is the stronger all-rounder.GPT-4.1 mini is cheaper.
Scores updated · 37 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. 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 software04GPT-54 of 4 tests
- ReasoningHard problems that need careful thinking04GPT-54 of 4 tests
- Images and chartsUnderstanding pictures, charts and video04GPT-54 of 4 tests
- FactsGetting facts right instead of making them up03GPT-53 of 3 tests
- AgentsCarrying out multi-step tasks on its own02GPT-52 of 2 tests
- Following instructionsDoing exactly what it is asked02GPT-52 of 2 tests
- MathCompetition and research-level math01GPT-51 of 1 test
- Long documentsFinding answers in very long texts01GPT-51 of 1 test
Math 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 82% 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-5 pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+51.3points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+48.7points ahead
- Short factual questions, answered correctlySimpleQA Verified+37.4points ahead
Where GPT-4.1 mini pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 37 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5
- SWE-bench VerifiedGPT-5 by 51.323.674.9+51.3
- Terminal-Bench 2.1GPT-5 by 25.110.135.2+25.1
- Terminal-Bench HardGPT-5 by 257.632.6+25
- SciCodeGPT-5 by 2.540.442.9+2.5
ReasoningGPT-5
- Humanity's Last ExamGPT-5 by 23.5528.5+23.5
- GPQA DiamondGPT-5 by 1966.485.4+19
- ARC-AGI-2GPT-5 by 9.909.9+9.9
- CritPtGPT-5 by 5.705.7+5.7
FactsGPT-5
- SimpleQA VerifiedGPT-5 by 37.412.750.1+37.4
- AA-Omniscience · AccuracyGPT-5 by 2020.340.3+20
- AA-Omniscience · Non-hallucinationGPT-5 by 10.57.317.8+10.5
Images and chartsGPT-5
- CharXiv (RQ)GPT-5 by 24.356.881.1+24.3
- MMMU-ProGPT-5 by 15.558.774.2+15.5
- MMMUGPT-5 by 11.572.784.2+11.5
- LMArena · VisionGPT-5 by 27 rating points11811208+27 rating
Following instructionsGPT-5
- IFBenchGPT-5 by 34.838.373.1+34.8
- Multi-ChallengeGPT-5 by 33.835.869.6+33.8
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.
- ARC-AGI-1GPT-5 by 62.23.565.7+62.2
- HMMT 2025GPT-5 by 58.33593.3+58.3
- AIME 2025GPT-5 by 54.440.294.6+54.4
- Aider-PolyglotGPT-5 by 53.334.788+53.3
- AA-OmniscienceGPT-5 by 44.9-53.6-8.7+44.9
- swe_bench_bashGPT-5 by 41.123.965+41.1
- τ²-Bench Telecom (AA run)GPT-5 by 31.952.984.8+31.9
- AIR-Bench 2024GPT-5 by 27.360.487.7+27.3
- AA Agentic IndexGPT-5 by 24.71.826.5+24.7
- Artificial Analysis Coding IndexGPT-5 by 17.620.237.8+17.6
- AA IntelligenceGPT-5 by 12.810.223+12.8
- HarmBenchGPT-5 by 1285.697.6+12
- bbqGPT-5 by 4.792.196.8+4.7
- XSTesttie97.497.1tie
- anthropic_red_teamtie99.399.1tie
- simple_safety_teststie10099.8tie
Questions people ask
Which is better, GPT-4.1 mini or GPT-5?
GPT-5 wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. GPT-4.1 mini wins none, but costs 82% less.
Which is better for coding?
GPT-5. It wins 4 of the 4 coding tests both models report; GPT-4.1 mini wins none.
Which is cheaper?
GPT-4.1 mini costs $0.40 per million input tokens and $1.60 per million output tokens; GPT-5 costs $1.25 and $10.00. That makes GPT-4.1 mini about 82% cheaper for the same work.
How do you compare the two?
We use the 37 benchmark tests both models have published scores on. The verdict counts the 21 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.