GPT-5.2 vs GPT-5 mini
Wins 6 of 8 areas
Coding · Reasoning · Facts · Images and charts · Math · Long documents
Wins 0 of 8 areas
—
GPT-5.2 is the stronger all-rounder.GPT-5 mini is cheaper.
Scores updated · 35 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.
- ReasoningHard problems that need careful thinking40GPT-5.24 of 4 tests
- CodingWriting and fixing software30GPT-5.23 of 3 tests
- FactsGetting facts right instead of making them up31GPT-5.23 of 4 tests
- Images and chartsUnderstanding pictures, charts and video20GPT-5.22 of 2 tests
- MathCompetition and research-level math20GPT-5.22 of 2 tests
- Long documentsFinding answers in very long texts10GPT-5.21 of 1 test
- AgentsCarrying out multi-step tasks on its own11Even1 each
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Long documents and following instructions 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-5 mini costs 86% 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-5.2 pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+48.5points ahead
- Real work tasks from 44 professionsGDPVal+34.8points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+20.7points ahead
Where GPT-5 mini pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+24.8points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+4.4points 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-5.2
- SWE-bench VerifiedGPT-5.2 by 20.28059.8+20.2
- Terminal-Bench HardGPT-5.2 by 13.74733.3+13.7
- SciCodeGPT-5.2 by 13.152.139+13.1
AgentsEven
- GDPValGPT-5.2 by 34.848.313.5+34.8
- τ-Bench V3 · BankingGPT-5 mini by 4.411.115.5+4.4
ReasoningGPT-5.2
- ARC-AGI-2GPT-5.2 by 48.552.94.4+48.5
- Humanity's Last ExamGPT-5.2 by 16.237.721.5+16.2
- CritPtGPT-5.2 by 11.611.60+11.6
- GPQA DiamondGPT-5.2 by 7.590.382.8+7.5
FactsGPT-5.2
- AA-Omniscience · Non-hallucinationGPT-5 mini by 24.818.843.6+24.8
- AA-Omniscience · AccuracyGPT-5.2 by 19.344.325+19.3
- SimpleQA VerifiedGPT-5.2 by 15.537.121.6+15.5
- Vectara HHEM hallucination ratelower is betterGPT-5.2 by 2.110.812.9+2.1
Images and chartsGPT-5.2
- MMMU-ProGPT-5.2 by 4.574.670.1+4.5
- LMArena · VisionGPT-5.2 by 44 rating points12461202+44 rating
MathGPT-5.2
- FrontierMath Tiers 1-3 (v2)GPT-5.2 by 20.767.446.7+20.7
- FrontierMath Tier 4GPT-5.2 by 19.531.712.2+19.5
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 Agentic IndexGPT-5.2 by 51.360.28.9+51.3
- Artificial Analysis Coding IndexGPT-5.2 by 33.148.715.6+33.1
- ARC-AGI-1GPT-5.2 by 31.986.254.3+31.9
- FrontierMath (overall)GPT-5.2 by 18.240.322.1+18.2
- AA-OmniscienceGPT-5.2 by 16.4-0.9-17.3+16.4
- τ²-Bench Telecom (AA run)GPT-5.2 by 16.484.868.4+16.4
- AA IntelligenceGPT-5.2 by 13.630.416.8+13.6
- frontiermath_tier_4_v1GPT-5.2 by 12.518.86.3+12.5
- HMMT 2025GPT-5.2 by 11.699.487.8+11.6
- HMMT Nov. 2025GPT-5.2 by 11.695.884.2+11.6
- swe_bench_bashGPT-5.2 by 9.26959.8+9.2
- AIME 2025GPT-5.2 by 8.910091.1+8.9
- matharena_visual_math_overallGPT-5.2 by 8.386.578.2+8.3
- vectara_factual_consistencyGPT-5.2 by 2.189.287.1+2.1
- vectara_avg_summary_lengthGPT-5.2 by 16.6 rating points186.3169.7+16.6 rating
- vectara_answer_ratetie10099.9tie
Questions people ask
Which is better, GPT-5.2 or GPT-5 mini?
GPT-5.2 wins six of the eight areas we test: coding, reasoning, facts, images and charts, math and long documents. GPT-5 mini wins none, but costs 86% less. They are level on agents and following instructions.
Which is better for coding?
GPT-5.2. It wins 3 of the 3 coding tests both models report; GPT-5 mini wins none.
Which is cheaper?
GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens; GPT-5 mini costs $0.25 and $2.00. That makes GPT-5 mini about 86% 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.