GPT-5.2 vs GPT-5 Nano
Updated October 21, 2026. 82 benchmarks have a result for both. Coding is ranked on the Vibe Coding Index; every other board is shown as published.
Decision reading
Pick GPT-5.2 for coding: it scores 58.9 on the Vibe Coding Index against 34.8 for GPT-5 Nano.
Pick GPT-5 Nano when output tokens cost 35.0x less, or when a solved task costs $0.109 against $1.91.
- Shared results
- 82
- GPT-5.2 only
- 29
- GPT-5 Nano only
- 3
- Like-for-like categories
- 8 of 9
Shared results can support a head-to-head reading. Results only one model has describe coverage, not superiority.
Which one for your work
Each card is decided by one number both models carry. Where only one has it, the card says so and makes no call.
Coding
Vibe Coding Index, four public coding benchmarks
GPT-5.2
GPT-5.2 scores 58.9 against 34.8.
2 and 1 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-5.2
GPT-5.2 leads, 30.4 against 13.0.
Same board, both measured
Reasoning
GPQA Diamond
GPT-5.2
GPT-5.2 leads, 91.7% against 63.4%.
Same board, both measured
Math
AIME
GPT-5.2
GPT-5.2 leads, 96.9% against 81.2%.
Same board, both measured
Knowledge
MMLU Pro
GPT-5.2
GPT-5.2 leads, 86.2% against 76.1%.
Same board, both measured
Web design
Design Arena: websites
GPT-5.2
GPT-5.2 leads, 1196 against 1101.
Same board, both measured
Long documents
Listed context window
No call
Both take 400K tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-5 Nano
GPT-5.2 $0.0088, GPT-5 Nano $0.0003 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
GPT-5 Nano
GPT-5.2 $1.91, GPT-5 Nano $0.109.
Published attempt cost and pass rate
At a glance
- Coding index
- GPT-5.2 58.9 from 2 benchmarks, GPT-5 Nano 34.8 from 1 benchmark.
- Widest category
- Agentic work, 48.2 points apart on average over 1 shared benchmark.
- Widest single gap
- ARC-AGI: GPT-5.2 86.2% against GPT-5 Nano 20.7%, 65.5 points apart on a 0 to 100 scale.
- Output price
- GPT-5.2 $14.00 against GPT-5 Nano $0.40 per million output tokens, 35.0x apart.
- Context window
- Both take 400K tokens.
Benchmark by benchmark
First the score: boards each model leads, per category, where both have a result. Then every board, a category to a fold; bars are each result on a 0 to 100 scale, so compare the two in a row, not one row with another.
| Category | GPT-5.2 | Bars | GPT-5 Nano | Ties |
|---|---|---|---|---|
| Coding | 8 | 0 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 1 | 0 | ||
| Reasoning | 15 | 0 | ||
| Math | 10 | 0 | ||
| Knowledge | 13 | 0 | ||
| Web design | 6 | 0 | ||
| General | 27 | 0 | ||
| All shared boards | 82 | 0 |
CodingLike-for-likeFour of these boards feed the Vibe Coding Index80
| Benchmark | GPT-5.2 | GPT-5 Nano | Leader |
|---|---|---|---|
| SWE-bench Verified | 72.8% | 34.8% | GPT-5.2 |
| SWE-bench Multilingual | 66.7% | Not measured | Only GPT-5.2 measured |
| LMArena WebDev | 1417 | Not measured | Only GPT-5.2 measured |
| LiveCodeBench (Vals AI) | 85.4% | 70.2% | GPT-5.2 |
| SWE-bench Verified (Vals AI) | 75.8% | Not measured | Only GPT-5.2 measured |
| Vibe Code Bench (Vals AI) | 53.5% | Not measured | Only GPT-5.2 measured |
| LiveCodeBench (Artificial Analysis run) | 89.4% | 78.9% | GPT-5.2 |
| Terminal-Bench Hard (Artificial Analysis run) | 47.0% | 17.4% | GPT-5.2 |
| Terminal-Bench | 64.9% | 21.8% | GPT-5.2 |
| GSO-Bench | 27.5% | Not measured | Only GPT-5.2 measured |
| WeirdML | 72.2% | 38.1% | GPT-5.2 |
| ALE-Bench | 1293.5 | Not measured | Only GPT-5.2 measured |
| LMArena Text: coding | 1442 | 1351 | GPT-5.2 |
| LMArena Text: software and IT | 1434 | 1345 | GPT-5.2 |
| LMArena WebDev: HTML | 1427 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 23 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5.2 | GPT-5 Nano | Leader |
|---|---|---|---|
| AA Intelligence Index | 30.4 | 13.0 | GPT-5.2 |
| Epoch Capabilities Index | 153.7 | 139.2 | GPT-5.2 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index10
| Benchmark | GPT-5.2 | GPT-5 Nano | Leader |
|---|---|---|---|
| Design Arena: agents building web apps | 1086 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building full-stack apps | 1043 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building mobile apps | 1094 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Android apps | 1064 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Godot games | 1142 | Not measured | Only GPT-5.2 measured |
| τ²-Bench Telecom | 84.8% | 36.5% | GPT-5.2 |
| METR time horizons | 75.3% | Not measured | Only GPT-5.2 measured |
| Remote Labor Index | 2.5% | Not measured | Only GPT-5.2 measured |
| GDPval | 49.7% | Not measured | Only GPT-5.2 measured |
| DeepResearch Bench | 41.1% | Not measured | Only GPT-5.2 measured |
| Vending-Bench 2 | $3,591 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 28 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index150
| Benchmark | GPT-5.2 | GPT-5 Nano | Leader |
|---|---|---|---|
| GPQA Diamond | 91.7% | 63.4% | GPT-5.2 |
| MMMU Pro (Vals AI) | 86.7% | 70.9% | GPT-5.2 |
| GPQA Diamond (Artificial Analysis run) | 90.3% | 67.6% | GPT-5.2 |
| Humanity's Last Exam (Artificial Analysis run) | 37.7% | 9.5% | GPT-5.2 |
| GPQA Diamond (Epoch AI run) | 91.4% | 69.4% | GPT-5.2 |
| ARC-AGI | 86.2% | 20.7% | GPT-5.2 |
| ARC-AGI-2 | 52.9% | 2.6% | GPT-5.2 |
| SimpleBench | 45.8% | Not measured | Only GPT-5.2 measured |
| Chess puzzles | 48.0% | 22.0% | GPT-5.2 |
| Mystery game puzzles | 23.0% | 9.0% | GPT-5.2 |
| EBR-bench | 23.0% | Not measured | Only GPT-5.2 measured |
| VPCT | 84.0% | 37.2% | GPT-5.2 |
| LMArena Text: hard prompts | 1428 | 1328 | GPT-5.2 |
| LMArena Text: expert | 1445 | 1321 | GPT-5.2 |
| LMArena Text: hard prompts in English | 1431 | 1338 | GPT-5.2 |
| LMArena Vision: diagrams | 1265 | 1141 | GPT-5.2 |
| LMArena Vision: homework | 1276 | 1157 | GPT-5.2 |
| Neither model has a result on the other 4 boards here | |||
MathLike-for-likePublic boards, shown as published, kept out of the index100
| Benchmark | GPT-5.2 | GPT-5 Nano | Leader |
|---|---|---|---|
| AIME | 96.9% | 81.2% | GPT-5.2 |
| MATH 500 (Vals AI) | Not measured | 93.8% | Only GPT-5 Nano measured |
| MGSM (Vals AI) | 94.0% | 89.3% | GPT-5.2 |
| AA Math Index | 99.0 | 83.7 | GPT-5.2 |
| AIME 2025 (Artificial Analysis run) | 99.0% | 83.7% | GPT-5.2 |
| OTIS Mock AIME | 96.1% | 81.1% | GPT-5.2 |
| FrontierMath, tiers 1 to 3 | 67.4% | 20.0% | GPT-5.2 |
| FrontierMath, tier 4 | 31.7% | 2.4% | GPT-5.2 |
| ProofBench | 15.0% | 12.0% | GPT-5.2 |
| MATH level 5 | Not measured | 95.2% | Only GPT-5 Nano measured |
| LMArena Text: math | 1440 | 1317 | GPT-5.2 |
| LMArena Text: mathematical work | 1442 | 1326 | GPT-5.2 |
| Neither model has a result on the other 1 board here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index130
| Benchmark | GPT-5.2 | GPT-5 Nano | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | 76.1% | GPT-5.2 |
| LegalBench (Vals AI) | 82.8% | 50.1% | GPT-5.2 |
| CaseLaw (Vals AI) | 66.0% | 52.6% | GPT-5.2 |
| CorpFin (Vals AI) | 65.9% | Not measured | Only GPT-5.2 measured |
| TaxEval (Vals AI) | 75.8% | 67.4% | GPT-5.2 |
| MortgageTax (Vals AI) | 67.1% | 53.6% | GPT-5.2 |
| MedQA (Vals AI) | 94.1% | 93.3% | GPT-5.2 |
| MedCode (Vals AI) | 49.8% | 30.4% | GPT-5.2 |
| MMLU Pro (Artificial Analysis run) | 87.4% | 78.0% | GPT-5.2 |
| SimpleQA Verified | 37.1% | 11.7% | GPT-5.2 |
| LMArena Text: science | 1425 | 1318 | GPT-5.2 |
| LMArena Text: medicine and healthcare | 1425 | 1335 | GPT-5.2 |
| LMArena Text: legal and government | 1427 | 1298 | GPT-5.2 |
| LMArena Text: business and finance | 1414 | 1311 | GPT-5.2 |
| Neither model has a result on the other 3 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index60
| Benchmark | GPT-5.2 | GPT-5 Nano | Leader |
|---|---|---|---|
| Design Arena: websites | 1196 | 1101 | GPT-5.2 |
| Design Arena: UI components | 1190 | 1066 | GPT-5.2 |
| Design Arena: game dev | 1196 | 1045 | GPT-5.2 |
| Design Arena: data viz | 1202 | 1056 | GPT-5.2 |
| Design Arena: 3D | 1079 | 966 | GPT-5.2 |
| Design Arena: all code categories | 1171 | 1084 | GPT-5.2 |
| Design Arena: SVG | 1137 | Not measured | Only GPT-5.2 measured |
| Design Arena: ASCII art | 1153 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 1 board here | |||
GeneralLike-for-likePublic boards, shown as published, kept out of the index270
| Benchmark | GPT-5.2 | GPT-5 Nano | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 49.3% | 30.4% | GPT-5.2 |
| MedScribe (Vals AI) | 84.4% | 72.9% | GPT-5.2 |
| IFBench | 75.4% | 67.6% | GPT-5.2 |
| Long context reasoning (AA-LCR) | 82.7% | 45.0% | GPT-5.2 |
| DTBench | 90.9% | 62.7% | GPT-5.2 |
| LMCA | 43.9% | 7.9% | GPT-5.2 |
| Fiction.LiveBench | Not measured | 44.4% | Only GPT-5 Nano measured |
| CL-bench | 18.2% | Not measured | Only GPT-5.2 measured |
| Furniture assembly | 38.3% | Not measured | Only GPT-5.2 measured |
| LMArena Text | 1416 | 1320 | GPT-5.2 |
| LMArena Text: instruction following | 1409 | 1306 | GPT-5.2 |
| LMArena Text: creative writing | 1376 | 1249 | GPT-5.2 |
| LMArena Vision | 1246 | 1159 | GPT-5.2 |
| LMArena Text: English | 1422 | 1321 | GPT-5.2 |
| LMArena Text: non-English | 1405 | 1313 | GPT-5.2 |
| LMArena Text: Chinese | 1447 | 1356 | GPT-5.2 |
| LMArena Text: German | 1444 | 1327 | GPT-5.2 |
| LMArena Text: Japanese | 1420 | 1227 | GPT-5.2 |
| LMArena Text: Korean | 1374 | 1269 | GPT-5.2 |
| LMArena Text: French | 1454 | Not measured | Only GPT-5.2 measured |
| LMArena Text: Spanish | 1413 | 1360 | GPT-5.2 |
| LMArena Text: Russian | 1415 | 1296 | GPT-5.2 |
| LMArena Text: Polish | 1426 | 1314 | GPT-5.2 |
| LMArena Text: multi-turn | 1419 | 1311 | GPT-5.2 |
| LMArena Text: longer queries | 1413 | 1312 | GPT-5.2 |
| LMArena Text: writing and language | 1392 | 1283 | GPT-5.2 |
| LMArena Text: entertainment and media | 1378 | 1268 | GPT-5.2 |
| LMArena Vision: English | 1243 | 1170 | GPT-5.2 |
| LMArena Vision: Chinese | 1305 | 1176 | GPT-5.2 |
| LMArena Vision: reading text | 1257 | 1152 | GPT-5.2 |
| LMArena Vision: humor | 1227 | 1110 | GPT-5.2 |
| LMArena Vision: captioning | 1267 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: recognizing things | 1232 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: creative writing | 1225 | Not measured | Only GPT-5.2 measured |
| LMArena Document | 1405 | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 11 boards here | |||
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.2 | GPT-5 Nano | Leader |
|---|---|---|---|
| Neither model has a result on 5 boards here | |||
What each workload costs
Three fixed token mixes at each model's listed price, so per-token rates become a bill.
Chat turn
1K input + 500 output tokens
GPT-5 Nano costs less
Repository review
50K input + 3K output tokens
GPT-5 Nano costs less
Agent loop
200K input + 10K output tokens
GPT-5 Nano costs less
Listed OpenRouter rates, without cache or batch discounts. A workload that does not fit a model's context window would need splitting.
Specification differences
The leading value in a row is marked; a row with a missing value names no leader. Our two rows are marked in red.
| Specification | OpenAI GPT-5.2 | OpenAI GPT-5 Nano |
|---|---|---|
| Maker | OpenAI | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | December 10, 2025 | August 7, 2025 |
| Context window | 400K | 400K |
| Input $/Mtok | $1.75 | $0.05 |
| Output $/Mtok | $14.00 | $0.40 |
| Coding index | 58.9 | 34.8 |
| Independent measurements behind it | 2 | 1 |
| Cost per solved task | $1.91 | $0.109 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.2 better than GPT-5 Nano for coding?
Pick GPT-5.2 for coding: it scores 58.9 on the Vibe Coding Index against 34.8 for GPT-5 Nano. Pick GPT-5 Nano when output tokens cost 35.0x less, or when a solved task costs $0.109 against $1.91.
Which is cheaper, GPT-5.2 or GPT-5 Nano?
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. Per solved SWE-bench Verified task, GPT-5.2 costs $1.91 and GPT-5 Nano $0.109.
Which has the bigger context window, GPT-5.2 or GPT-5 Nano?
Neither: both take 400K tokens.
How many benchmarks is the GPT-5.2 vs GPT-5 Nano comparison based on?
82 benchmarks have a result for both models as of October 21, 2026. GPT-5.2's coding index rests on 2 independent benchmarks and GPT-5 Nano's on 1. Where only one model has a result, the page says so instead of naming a winner.