GPT-5.2 vs GPT-5.4 Mini
GPT-5.2 leads GPT-5.4 Mini on coding, 55.9 to 35.8 on the Vibe Coding Index, and GPT-5.4 Mini costs 3.1x less per output token ($4.50 vs $14.00 per million). Updated October 22, 2026; 75 benchmarks have a result for both.
GPT-5.2 or GPT-5.4 Mini: which to pick
Pick GPT-5.2 for coding: it scores 55.9 on the Vibe Coding Index against 35.8 for GPT-5.4 Mini, though on 3 benchmarks against 5.
Pick GPT-5.4 Mini for output tokens 3.1x cheaper ($4.50 vs $14.00 per million) and a lower cost per solved task ($0.70 against $1.91).
- Shared results
- 75
- GPT-5.2 only
- 39
- GPT-5.4 Mini only
- 14
- Like-for-like categories
- 7 of 9
Only benchmarks that tested both models count toward a pick; a result only one model has is shown but not compared.
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, six public coding benchmarks
GPT-5.2
GPT-5.2 scores 55.9 against 35.8.
3 and 5 benchmarks behind the scores
Intelligence
AA Intelligence Index
GPT-5.2
GPT-5.2 leads, 30.4 against 24.1.
Same board, both measured
Agentic work
AA Agentic Index
No call
Only GPT-5.4 Mini has a AA Agentic Index result, 17.9.
Directional only
Reasoning
GPQA Diamond
GPT-5.2
GPT-5.2 leads, 91.7% against 83.1%.
Same board, both measured
Math
AIME
GPT-5.2
GPT-5.2 leads, 96.9% against 95.6%.
Same board, both measured
Knowledge
MMLU Pro
GPT-5.2
GPT-5.2 leads, 86.2% against 84.5%.
Same board, both measured
Web design
Design Arena: websites
No call
Only GPT-5.2 has a Design Arena: websites result, 1195.
Directional only
Long documents
Listed context window
No call
Both take 400K tokens.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
GPT-5.4 Mini
GPT-5.2 $0.0088, GPT-5.4 Mini $0.0030 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
GPT-5.4 Mini
GPT-5.2 $1.91, GPT-5.4 Mini $0.70.
Published attempt cost and pass rate
At a glance
- Coding index
- GPT-5.2 55.9 from 3 benchmarks, GPT-5.4 Mini 35.8 from 5 benchmarks.
- Widest category
- Reasoning, 11.2 points apart on average over 15 shared benchmarks.
- Widest single gap
- Kagi LLM Benchmark: GPT-5.2 73.3% against GPT-5.4 Mini 37.9%, 35.4 points apart on a 0 to 100 scale.
- Output price
- GPT-5.2 $14.00 against GPT-5.4 Mini $4.50 per million output tokens, 3.1x 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-5.4 Mini | Ties |
|---|---|---|---|---|
| Coding | 8 | 2 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 2 | 0 | ||
| Reasoning | 13 | 2 | ||
| Math | 6 | 1 | ||
| Knowledge | 9 | 1 | ||
| General | 23 | 6 | ||
| All shared boards | 63 | 12 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index82
| Benchmark | GPT-5.2 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| SWE-bench Verified | 72.8% | Not measured | Only GPT-5.2 measured |
| SWE-bench Multilingual | 66.7% | Not measured | Only GPT-5.2 measured |
| LMArena WebDev | 1417 | 1397 | GPT-5.2 |
| AA Coding Index | Not measured | 56.1 | Only GPT-5.4 Mini measured |
| LiveCodeBench (Vals AI) | 85.4% | 81.5% | GPT-5.2 |
| SWE-bench Verified (Vals AI) | 75.8% | 73.0% | GPT-5.2 |
| Terminal-Bench 4 (Vals AI) | Not measured | 2.5% | Only GPT-5.4 Mini measured |
| Vibe Code Bench (Vals AI) | 53.5% | 48.0% | GPT-5.2 |
| Code migration (Vals AI) | Not measured | 12.9% | Only GPT-5.4 Mini measured |
| ProgramBench (Vals AI) | Not measured | 0.0% | Only GPT-5.4 Mini measured |
| Terminal-Bench 2.1 (Vals AI) | Not measured | 54.7% | Only GPT-5.4 Mini measured |
| LiveCodeBench (Artificial Analysis run) | 89.4% | Not measured | Only GPT-5.2 measured |
| SciCode | Not measured | 52.1% | Only GPT-5.4 Mini measured |
| Terminal-Bench Hard (Artificial Analysis run) | 47.0% | 52.3% | GPT-5.4 Mini |
| Terminal-Bench | 64.9% | Not measured | Only GPT-5.2 measured |
| FrontierCode | Not measured | 27.0% | Only GPT-5.4 Mini measured |
| GSO-Bench | 27.5% | Not measured | Only GPT-5.2 measured |
| WeirdML | 72.2% | 60.3% | GPT-5.2 |
| ALE-Bench | 1293.5 | 1188.6 | GPT-5.2 |
| LMArena Text: coding | 1442 | 1438 | GPT-5.2 |
| LMArena Text: software and IT | 1434 | 1436 | GPT-5.4 Mini |
| LMArena WebDev: React | Not measured | 1386 | Only GPT-5.4 Mini measured |
| LMArena WebDev: HTML | 1427 | 1419 | GPT-5.2 |
| Neither model has a result on the other 16 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5.2 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| AA Intelligence Index | 30.4 | 24.1 | GPT-5.2 |
| Epoch Capabilities Index | 153.7 | 148.7 | GPT-5.2 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5.2 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| AA Agentic Index | Not measured | 17.9 | Only GPT-5.4 Mini measured |
| 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 | 1093 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Android apps | 1059 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Godot games | 1142 | Not measured | Only GPT-5.2 measured |
| Finance Agent (Vals AI) | Not measured | 45.4% | Only GPT-5.4 Mini measured |
| Tax Agent Bench (Vals AI) | Not measured | 8.8% | Only GPT-5.4 Mini measured |
| Legal Research Bench (Vals AI) | Not measured | 12.5% | Only GPT-5.4 Mini measured |
| Excel Modeling Benchmark (Vals AI) | Not measured | 45.4% | Only GPT-5.4 Mini measured |
| τ²-Bench Telecom | 84.8% | 83.3% | 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% | 36.3% | GPT-5.2 |
| Vending-Bench 2 | $3,591 | Not measured | Only GPT-5.2 measured |
| Berkeley Function Calling Leaderboard (BFCL) | 55.9% | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 24 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index132
| Benchmark | GPT-5.2 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| GPQA Diamond | 91.7% | 83.1% | GPT-5.2 |
| MMMU Pro (Vals AI) | 86.7% | 79.3% | GPT-5.2 |
| GPQA Diamond (Artificial Analysis run) | 90.3% | 87.5% | GPT-5.2 |
| Humanity's Last Exam (Artificial Analysis run) | 37.7% | 28.1% | GPT-5.2 |
| GPQA Diamond (Epoch AI run) | 91.4% | 86.9% | GPT-5.2 |
| ARC-AGI | 86.2% | 63.7% | GPT-5.2 |
| ARC-AGI-2 | 52.9% | 18.9% | GPT-5.2 |
| SimpleBench | 45.8% | Not measured | Only GPT-5.2 measured |
| Chess puzzles | 48.0% | 23.0% | GPT-5.2 |
| Mystery game puzzles | 23.0% | 11.0% | GPT-5.2 |
| EBR-bench | 23.0% | Not measured | Only GPT-5.2 measured |
| VPCT | 84.0% | Not measured | Only GPT-5.2 measured |
| LMArena Text: hard prompts | 1428 | 1424 | GPT-5.2 |
| LMArena Text: expert | 1445 | 1435 | GPT-5.2 |
| LMArena Text: hard prompts in English | 1431 | 1420 | GPT-5.2 |
| LMArena Vision: diagrams | 1265 | 1266 | GPT-5.4 Mini |
| LMArena Vision: homework | 1276 | 1285 | GPT-5.4 Mini |
| Kagi LLM Benchmark | 73.3% | 37.9% | 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 index61
| Benchmark | GPT-5.2 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| AIME | 96.9% | 95.6% | GPT-5.2 |
| MGSM (Vals AI) | 94.0% | Not measured | Only GPT-5.2 measured |
| AA Math Index | 99.0 | Not measured | Only GPT-5.2 measured |
| AIME 2025 (Artificial Analysis run) | 99.0% | Not measured | Only GPT-5.2 measured |
| OTIS Mock AIME | 96.1% | 88.9% | GPT-5.2 |
| FrontierMath, tiers 1 to 3 | 67.4% | 51.2% | GPT-5.2 |
| FrontierMath, tier 4 | 31.7% | 9.8% | GPT-5.2 |
| ProofBench | 15.0% | 21.0% | GPT-5.4 Mini |
| LMArena Text: math | 1440 | 1418 | GPT-5.2 |
| LMArena Text: mathematical work | 1442 | 1429 | GPT-5.2 |
| Neither model has a result on the other 3 boards here | |||
KnowledgeLike-for-likePublic boards, shown as published, kept out of the index91
| Benchmark | GPT-5.2 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | 84.5% | GPT-5.2 |
| LegalBench (Vals AI) | 82.8% | Not measured | Only GPT-5.2 measured |
| CaseLaw (Vals AI) | 66.0% | 51.7% | GPT-5.2 |
| CorpFin (Vals AI) | 65.9% | 60.9% | GPT-5.2 |
| TaxEval (Vals AI) | 75.8% | 71.2% | GPT-5.2 |
| MortgageTax (Vals AI) | 67.1% | 63.5% | GPT-5.2 |
| MedQA (Vals AI) | 94.1% | Not measured | Only GPT-5.2 measured |
| MedCode (Vals AI) | 49.8% | Not measured | Only GPT-5.2 measured |
| MMLU Pro (Artificial Analysis run) | 87.4% | Not measured | Only GPT-5.2 measured |
| SimpleQA Verified | 37.1% | 29.4% | GPT-5.2 |
| LMArena Text: science | 1425 | 1417 | GPT-5.2 |
| LMArena Text: medicine and healthcare | 1425 | 1395 | GPT-5.2 |
| LMArena Text: legal and government | 1427 | 1414 | GPT-5.2 |
| LMArena Text: business and finance | 1414 | 1418 | GPT-5.4 Mini |
| FACTS Benchmark Suite | 48.5% | Not measured | Only GPT-5.2 measured |
| Neither model has a result on the other 3 boards here | |||
Web designDirectional onlyPublic boards, shown as published, kept out of the index8 boards, none shared
| Benchmark | GPT-5.2 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| Design Arena: websites | 1195 | Not measured | Only GPT-5.2 measured |
| Design Arena: UI components | 1190 | Not measured | Only GPT-5.2 measured |
| Design Arena: game dev | 1195 | Not measured | Only GPT-5.2 measured |
| Design Arena: data viz | 1202 | Not measured | Only GPT-5.2 measured |
| Design Arena: 3D | 1079 | Not measured | Only GPT-5.2 measured |
| Design Arena: all code categories | 1170 | Not measured | Only GPT-5.2 measured |
| Design Arena: SVG | 1139 | Not measured | Only GPT-5.2 measured |
| Design Arena: ASCII art | 1152 | 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 index236
| Benchmark | GPT-5.2 | GPT-5.4 Mini | Leader |
|---|---|---|---|
| SAGE (Vals AI) | 49.3% | 50.8% | GPT-5.4 Mini |
| MedScribe (Vals AI) | 84.4% | Not measured | Only GPT-5.2 measured |
| Vals Multimodal Index | Not measured | 54.2% | Only GPT-5.4 Mini measured |
| IFBench | 75.4% | 73.3% | GPT-5.2 |
| Long context reasoning (AA-LCR) | 82.7% | 77.0% | GPT-5.2 |
| DTBench | 90.9% | 80.0% | GPT-5.2 |
| LMCA | 43.9% | 40.8% | GPT-5.2 |
| 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 | 1412 | GPT-5.2 |
| LMArena Text: instruction following | 1409 | 1405 | GPT-5.2 |
| LMArena Text: creative writing | 1376 | 1370 | GPT-5.2 |
| LMArena Vision | 1246 | 1245 | GPT-5.2 |
| LMArena Text: English | 1422 | 1412 | GPT-5.2 |
| LMArena Text: non-English | 1405 | 1405 | GPT-5.2 |
| LMArena Text: Chinese | 1447 | 1446 | GPT-5.2 |
| LMArena Text: German | 1444 | 1408 | GPT-5.2 |
| LMArena Text: Japanese | 1420 | 1375 | GPT-5.2 |
| LMArena Text: Korean | 1374 | 1370 | GPT-5.2 |
| LMArena Text: French | 1454 | 1441 | GPT-5.2 |
| LMArena Text: Spanish | 1413 | 1405 | GPT-5.2 |
| LMArena Text: Russian | 1415 | 1417 | GPT-5.4 Mini |
| LMArena Text: Polish | 1426 | 1415 | GPT-5.2 |
| LMArena Text: multi-turn | 1419 | 1429 | GPT-5.4 Mini |
| LMArena Text: longer queries | 1413 | 1408 | GPT-5.2 |
| LMArena Text: writing and language | 1392 | 1393 | GPT-5.4 Mini |
| LMArena Text: entertainment and media | 1378 | 1373 | GPT-5.2 |
| LMArena Vision: English | 1243 | 1244 | GPT-5.4 Mini |
| LMArena Vision: Chinese | 1305 | 1286 | GPT-5.2 |
| LMArena Vision: reading text | 1257 | 1253 | GPT-5.2 |
| LMArena Vision: humor | 1227 | 1235 | GPT-5.4 Mini |
| LMArena Vision: captioning | 1267 | Not measured | Only GPT-5.2 measured |
| LMArena Vision: recognizing things | 1232 | 1218 | GPT-5.2 |
| LMArena Vision: creative writing | 1225 | 1222 | GPT-5.2 |
| 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-5.4 Mini | 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-5.4 Mini costs less
Repository review
50K input + 3K output tokens
GPT-5.4 Mini costs less
Agent loop
200K input + 10K output tokens
GPT-5.4 Mini 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-5.4 Mini |
|---|---|---|
| Maker | OpenAI | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | December 10, 2025 | March 17, 2026 |
| Context window | 400K | 400K |
| Input $/Mtok | $1.75 | $0.75 |
| Output $/Mtok | $14.00 | $4.50 |
| Coding index | 55.9 | 35.8 |
| Independent measurements behind it | 3 | 5 |
| Cost per solved task | $1.91 | $0.70 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.2 better than GPT-5.4 Mini for coding?
Pick GPT-5.2 for coding: it scores 55.9 on the Vibe Coding Index against 35.8 for GPT-5.4 Mini, though on 3 benchmarks against 5. Pick GPT-5.4 Mini for output tokens 3.1x cheaper ($4.50 vs $14.00 per million) and a lower cost per solved task ($0.70 against $1.91).
Which is cheaper, GPT-5.2 or GPT-5.4 Mini?
GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens; GPT-5.4 Mini costs $0.75 and $4.50. Per solved SWE-bench Verified task, GPT-5.2 costs $1.91 and GPT-5.4 Mini $0.70.
Which has the bigger context window, GPT-5.2 or GPT-5.4 Mini?
Neither: both take 400K tokens.
How many benchmarks is the GPT-5.2 vs GPT-5.4 Mini comparison based on?
75 benchmarks have a result for both models as of October 22, 2026. GPT-5.2's coding index rests on 3 independent benchmarks and GPT-5.4 Mini's on 5. Where only one model has a result, the page says so instead of naming a winner.