Gemini 2.5 Pro vs GPT-5 mini
Wins 3 of 8 areas
Coding · Facts · Images and charts
Wins 4 of 8 areas
Agents · Math · Long documents · Following instructions
GPT-5 mini wins more areas, narrowly.Gemini 2.5 Pro is better at coding.
Scores updated · 42 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.
- AgentsCarrying out multi-step tasks on its own02GPT-5 mini2 of 3 tests · 1 tie
- MathCompetition and research-level math02GPT-5 mini2 of 2 tests
- Following instructionsDoing exactly what it is asked02GPT-5 mini2 of 2 tests
- Long documentsFinding answers in very long texts01GPT-5 mini1 of 1 test
- CodingWriting and fixing software31Gemini 2.5 Pro3 of 4 tests
- FactsGetting facts right instead of making them up31Gemini 2.5 Pro3 of 4 tests
- Images and chartsUnderstanding pictures, charts and video20Gemini 2.5 Pro2 of 2 tests
- ReasoningHard problems that need careful thinking11Even1 each · 2 ties
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-5 mini costs 80% 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 mini pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+34.5points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+26.7points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+22.1points ahead
Where Gemini 2.5 Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+34.4points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+24.8points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+14.1points ahead
Every test, side by side
All 42 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 2.5 Pro
- Terminal-Bench 2.1Gemini 2.5 Pro by 24.828.53.7+24.8
- SciCodeGemini 2.5 Pro by 7.346.339+7.3
- Terminal-Bench HardGPT-5 mini by 6.826.533.3+6.8
- SWE-bench VerifiedGemini 2.5 Pro by 3.463.259.8+3.4
AgentsGPT-5 mini
- GDPValGPT-5 mini by 13.5013.5+13.5
- τ-Bench V3 · BankingGPT-5 mini by 5.89.715.5+5.8
- Terminal-Bench 4.0tie00tie
ReasoningEven
- Humanity's Last ExamGPT-5 mini by 3.51821.5+3.5
- CritPtGemini 2.5 Pro by 2.62.60+2.6
- GPQA Diamondtie83.682.8tie
- ARC-AGI-2tie4.94.4tie
FactsGemini 2.5 Pro
- AA-Omniscience · Non-hallucinationGPT-5 mini by 34.59.143.6+34.5
- SimpleQA VerifiedGemini 2.5 Pro by 34.45621.6+34.4
- AA-Omniscience · AccuracyGemini 2.5 Pro by 14.13925+14.1
- Vectara HHEM hallucination ratelower is betterGemini 2.5 Pro by 5.9712.9+5.9
Images and chartsGemini 2.5 Pro
- LMArena · VisionGemini 2.5 Pro by 61 rating points12631202+61 rating
- MMMU-ProGemini 2.5 Pro by 4.874.970.1+4.8
MathGPT-5 mini
- FrontierMath Tiers 1-3 (v2)GPT-5 mini by 22.124.646.7+22.1
- FrontierMath Tier 4GPT-5 mini by 12.2012.2+12.2
Following instructionsGPT-5 mini
- IFBenchGPT-5 mini by 26.748.775.4+26.7
- Multi-ChallengeGPT-5 mini by 5.453.659+5.4
Other results20 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.
- HarmBenchGPT-5 mini by 31.765.497.1+31.7
- Artificial Analysis Coding IndexGemini 2.5 Pro by 31.146.715.6+31.1
- HMMT 2025GPT-5 mini by 22.165.787.8+22.1
- ARC-AGI-1GPT-5 mini by 17.33754.3+17.3
- τ²-Bench Telecom (AA run)GPT-5 mini by 14.354.168.4+14.3
- AIR-Bench 2024GPT-5 mini by 12.173.685.7+12.1
- AIME 2025GPT-5 mini by 8.18391.1+8.1
- vectara_avg_summary_lengthGPT-5 mini by 63.3 rating points106.4169.7+63.3 rating
- vectara_factual_consistencyGemini 2.5 Pro by 5.99387.1+5.9
- AA Agentic IndexGPT-5 mini by 5.43.58.9+5.4
- HMMT Nov. 2025GPT-5 mini by 4.28084.2+4.2
- simple_safety_testsGPT-5 mini by 397100+3
- frontiermath_tier_4_v1GPT-5 mini by 2.14.26.3+2.1
- AA IntelligenceGPT-5 mini by 1.81516.8+1.8
- AA-OmniscienceGemini 2.5 Pro by 1-16.4-17.3+1
- matharena_visual_math_overallGPT-5 mini by 177.278.2+1
- XSTestGemini 2.5 Pro by 198.797.7+1
- vectara_answer_ratetie99.199.9tie
- anthropic_red_teamtie99.599.1tie
- bbqtie96.496.3tie
Questions people ask
Which is better, Gemini 2.5 Pro or GPT-5 mini?
GPT-5 mini wins four of the eight areas we test: agents, math, long documents and following instructions. Gemini 2.5 Pro wins coding, facts and images and charts. They are level on reasoning.
Which is better for coding?
Gemini 2.5 Pro. It wins 3 of the 4 coding tests both models report; GPT-5 mini wins 1.
Which is cheaper?
Gemini 2.5 Pro costs $1.25 per million input tokens and $10.00 per million output tokens; GPT-5 mini costs $0.25 and $2.00. That makes GPT-5 mini about 80% cheaper for the same work.
How do you compare the two?
We use the 42 benchmark tests both models have published scores on. The verdict counts the 22 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.