GPT-5.2 vs GPT-5 nano
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Wins 0 of 8 areas
—
GPT-5.2 is the stronger all-rounder.GPT-5 nano is cheaper.
Scores updated · 33 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
- Images and chartsUnderstanding pictures, charts and video20GPT-5.22 of 2 tests
- MathCompetition and research-level math20GPT-5.22 of 2 tests
- FactsGetting facts right instead of making them up21GPT-5.22 of 4 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own10GPT-5.21 of 1 test
- Long documentsFinding answers in very long texts10GPT-5.21 of 1 test
- Following instructionsDoing exactly what it is asked10GPT-5.21 of 1 test
Agents, 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 nano costs 97% 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+50.3points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+47.4points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+45.2points ahead
Where GPT-5 nano pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+22.2points ahead
Every test, side by side
All 33 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 45.28034.8+45.2
- Terminal-Bench HardGPT-5.2 by 34.94712.1+34.9
- SciCodeGPT-5.2 by 15.552.136.6+15.5
ReasoningGPT-5.2
- ARC-AGI-2GPT-5.2 by 50.352.92.6+50.3
- Humanity's Last ExamGPT-5.2 by 28.237.79.5+28.2
- GPQA DiamondGPT-5.2 by 22.790.367.6+22.7
- CritPtGPT-5.2 by 11.611.60+11.6
FactsGPT-5.2
- SimpleQA VerifiedGPT-5.2 by 25.437.111.7+25.4
- AA-Omniscience · AccuracyGPT-5.2 by 25.244.319.1+25.2
- AA-Omniscience · Non-hallucinationGPT-5 nano by 22.218.841+22.2
- Vectara HHEM hallucination ratelower is bettertie10.810.5tie
Images and chartsGPT-5.2
- MMMU-ProGPT-5.2 by 13.674.661+13.6
- LMArena · VisionGPT-5.2 by 87 rating points12461159+87 rating
MathGPT-5.2
- FrontierMath Tiers 1-3 (v2)GPT-5.2 by 47.467.420+47.4
- FrontierMath Tier 4GPT-5.2 by 29.331.72.4+29.3
Other results15 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-5.2 by 69.586.216.7+69.5
- τ²-Bench Telecom (AA run)GPT-5.2 by 48.384.836.5+48.3
- AA Agentic IndexGPT-5.2 by 3760.223.2+37
- swe_bench_bashGPT-5.2 by 34.26934.8+34.2
- FrontierMath (overall)GPT-5.2 by 30.740.39.6+30.7
- Artificial Analysis Coding IndexGPT-5.2 by 28.448.720.3+28.4
- AA-OmniscienceGPT-5.2 by 27.8-0.9-28.7+27.8
- HMMT 2025GPT-5.2 by 23.899.475.6+23.8
- AA IntelligenceGPT-5.2 by 17.430.413+17.4
- frontiermath_tier_4_v1GPT-5.2 by 16.718.82.1+16.7
- AIME 2025GPT-5.2 by 14.810085.2+14.8
- HMMT Nov. 2025GPT-5.2 by 14.195.881.7+14.1
- vectara_avg_summary_lengthGPT-5.2 by 80.6 rating points186.3105.7+80.6 rating
- vectara_factual_consistencytie89.289.5tie
- vectara_answer_ratetie100100tie
Questions people ask
Which is better, GPT-5.2 or GPT-5 nano?
GPT-5.2 wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. GPT-5 nano wins none, but costs 97% less.
Which is better for coding?
GPT-5.2. It wins 3 of the 3 coding tests both models report; GPT-5 nano wins none.
Which is cheaper?
GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens; GPT-5 nano costs $0.05 and $0.40. That makes GPT-5 nano about 97% cheaper for the same work.
How do you compare the two?
We use the 33 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 15 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.