GPT-5.2 vs o3
GPT-5.2 leads o3 on the Artificial Analysis Intelligence Index, 30.4 to 20.2, and o3 costs 1.8x less per output token ($8.00 vs $14.00 per million). Updated October 21, 2026; 92 benchmarks have a result for both.
GPT-5.2 or o3: which to pick
Only GPT-5.2 has a coding score, 55.9 on the Vibe Coding Index; o3 has none, so the two cannot be ranked on coding.
Pick o3 for output tokens 1.8x cheaper ($8.00 vs $14.00 per million) and a lower cost per solved task ($0.571 against $1.91).
- Shared results
- 92
- GPT-5.2 only
- 22
- o3 only
- 10
- Like-for-like categories
- 8 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
No call
Only GPT-5.2 has a coding score, so no call is made.
One side not measured
Intelligence
AA Intelligence Index
GPT-5.2
GPT-5.2 leads, 30.4 against 20.2.
Same board, both measured
Reasoning
GPQA Diamond
GPT-5.2
GPT-5.2 leads, 91.7% against 84.1%.
Same board, both measured
Math
AIME
GPT-5.2
GPT-5.2 leads, 96.9% against 85.3%.
Same board, both measured
Knowledge
MMLU Pro
GPT-5.2
GPT-5.2 leads, 86.2% against 85.6%.
Same board, both measured
Web design
Design Arena: websites
GPT-5.2
GPT-5.2 leads, 1195 against 1036.
Same board, both measured
Long documents
Listed context window
GPT-5.2
GPT-5.2 takes 400K tokens against 200K.
Listed by OpenRouter
Everyday cost
1K input + 500 output tokens
o3
GPT-5.2 $0.0088, o3 $0.0060 per turn.
List prices
Cost per solved task
One SWE-bench Verified task, solved
o3
GPT-5.2 $1.91, o3 $0.571.
Published attempt cost and pass rate
At a glance
- Coding index
- GPT-5.2 55.9 from 3 benchmarks, o3 not benchmarked.
- Widest category
- Web design, 16.1 points apart on average over 5 shared benchmarks.
- Widest single gap
- ARC-AGI-2: GPT-5.2 52.9% against o3 6.5%, 46.4 points apart on a 0 to 100 scale.
- Output price
- GPT-5.2 $14.00 against o3 $8.00 per million output tokens, 1.8x apart.
- Context window
- GPT-5.2 takes 400K tokens, o3 200K.
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 | o3 | Ties |
|---|---|---|---|---|
| Coding | 9 | 0 | ||
| Intelligence | 2 | 0 | ||
| Agentic work | 3 | 2 | ||
| Reasoning | 15 | 2 | ||
| Math | 8 | 0 | ||
| Knowledge | 8 | 6 | ||
| Web design | 5 | 0 | ||
| General | 30 | 2 | ||
| All shared boards | 80 | 12 |
CodingLike-for-likeSix of these boards feed the Vibe Coding Index90
| Benchmark | GPT-5.2 | o3 | Leader |
|---|---|---|---|
| Aider polyglot | Not measured | 81.3% | Only o3 measured |
| SWE-bench Verified | 72.8% | 58.4% | GPT-5.2 |
| SWE-bench Multimodal | Not measured | 36.0% | Only o3 measured |
| 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% | 83.9% | 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% | 80.8% | GPT-5.2 |
| Terminal-Bench Hard (Artificial Analysis run) | 47.0% | 37.1% | GPT-5.2 |
| Terminal-Bench | 64.9% | Not measured | Only GPT-5.2 measured |
| GSO-Bench | 27.5% | 8.8% | GPT-5.2 |
| WeirdML | 72.2% | 52.4% | GPT-5.2 |
| ALE-Bench | 1293.5 | 933.5 | GPT-5.2 |
| LMArena Text: coding | 1442 | 1407 | GPT-5.2 |
| LMArena Text: software and IT | 1434 | 1415 | GPT-5.2 |
| LMArena WebDev: HTML | 1427 | Not measured | Only GPT-5.2 measured |
| SciCode | Not measured | 9.2% | Only o3 measured |
| Neither model has a result on the other 21 boards here | |||
IntelligenceLike-for-likePublic boards, shown as published, kept out of the index20
| Benchmark | GPT-5.2 | o3 | Leader |
|---|---|---|---|
| AA Intelligence Index | 30.4 | 20.2 | GPT-5.2 |
| Epoch Capabilities Index | 153.7 | 146.7 | GPT-5.2 |
Agentic workLike-for-likePublic boards, shown as published, kept out of the index32
| Benchmark | GPT-5.2 | o3 | 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 | 1093 | Not measured | Only GPT-5.2 measured |
| Design Arena: agents building Android apps | 1058 | 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% | 80.7% | GPT-5.2 |
| METR time horizons | 75.3% | 65.4% | GPT-5.2 |
| Remote Labor Index | 2.5% | Not measured | Only GPT-5.2 measured |
| GDPval | 49.7% | 30.8% | GPT-5.2 |
| DeepResearch Bench | 41.1% | 45.2% | o3 |
| OSWorld | Not measured | 23.0% | Only o3 measured |
| Vending-Bench 2 | $3,591 | Not measured | Only GPT-5.2 measured |
| Berkeley Function Calling Leaderboard (BFCL) | 55.9% | 63.0% | o3 |
| BrowseComp-Plus | Not measured | 65.9% | Only o3 measured |
| Neither model has a result on the other 27 boards here | |||
ReasoningLike-for-likePublic boards, shown as published, kept out of the index152
| Benchmark | GPT-5.2 | o3 | Leader |
|---|---|---|---|
| GPQA Diamond | 91.7% | 84.1% | GPT-5.2 |
| MMMU Pro (Vals AI) | 86.7% | 80.4% | GPT-5.2 |
| GPQA Diamond (Artificial Analysis run) | 90.3% | 82.7% | GPT-5.2 |
| Humanity's Last Exam (Artificial Analysis run) | 37.7% | 20.1% | GPT-5.2 |
| GPQA Diamond (Epoch AI run) | 91.4% | 81.8% | GPT-5.2 |
| ARC-AGI | 86.2% | 60.8% | GPT-5.2 |
| ARC-AGI-2 | 52.9% | 6.5% | GPT-5.2 |
| SimpleBench | 45.8% | 53.1% | o3 |
| Chess puzzles | 48.0% | 37.0% | GPT-5.2 |
| Mystery game puzzles | 23.0% | 29.0% | o3 |
| EBR-bench | 23.0% | Not measured | Only GPT-5.2 measured |
| VPCT | 84.0% | 52.0% | GPT-5.2 |
| LMArena Text: hard prompts | 1428 | 1402 | GPT-5.2 |
| LMArena Text: expert | 1445 | 1402 | GPT-5.2 |
| LMArena Text: hard prompts in English | 1431 | 1411 | GPT-5.2 |
| LMArena Vision: diagrams | 1265 | 1220 | GPT-5.2 |
| LMArena Vision: homework | 1276 | 1234 | GPT-5.2 |
| Kagi LLM Benchmark | 73.3% | 67.6% | 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 index80
| Benchmark | GPT-5.2 | o3 | Leader |
|---|---|---|---|
| AIME | 96.9% | 85.3% | GPT-5.2 |
| MATH 500 (Vals AI) | Not measured | 94.6% | Only o3 measured |
| MGSM (Vals AI) | 94.0% | 91.8% | GPT-5.2 |
| AA Math Index | 99.0 | 88.3 | GPT-5.2 |
| AIME 2025 (Artificial Analysis run) | 99.0% | 88.3% | GPT-5.2 |
| OTIS Mock AIME | 96.1% | 84.4% | GPT-5.2 |
| FrontierMath, tiers 1 to 3 | 67.4% | 33.3% | GPT-5.2 |
| FrontierMath, tier 4 | 31.7% | Not measured | Only GPT-5.2 measured |
| ProofBench | 15.0% | Not measured | Only GPT-5.2 measured |
| MATH level 5 | Not measured | 97.8% | Only o3 measured |
| LMArena Text: math | 1440 | 1426 | GPT-5.2 |
| LMArena Text: mathematical work | 1442 | 1423 | 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 index86
| Benchmark | GPT-5.2 | o3 | Leader |
|---|---|---|---|
| MMLU Pro | 86.2% | 85.6% | GPT-5.2 |
| LegalBench (Vals AI) | 82.8% | 83.8% | o3 |
| CaseLaw (Vals AI) | 66.0% | Not measured | Only GPT-5.2 measured |
| CorpFin (Vals AI) | 65.9% | 59.7% | GPT-5.2 |
| TaxEval (Vals AI) | 75.8% | 74.6% | GPT-5.2 |
| MortgageTax (Vals AI) | 67.1% | 65.7% | GPT-5.2 |
| MedQA (Vals AI) | 94.1% | 96.1% | o3 |
| MedCode (Vals AI) | 49.8% | 47.3% | GPT-5.2 |
| MMLU Pro (Artificial Analysis run) | 87.4% | 85.3% | GPT-5.2 |
| SimpleQA Verified | 37.1% | 49.4% | o3 |
| GeoBench | Not measured | 74.0% | Only o3 measured |
| LMArena Text: science | 1425 | 1429 | o3 |
| LMArena Text: medicine and healthcare | 1425 | 1448 | o3 |
| LMArena Text: legal and government | 1427 | 1428 | o3 |
| LMArena Text: business and finance | 1414 | 1393 | GPT-5.2 |
| FACTS Benchmark Suite | 48.5% | 44.4% | GPT-5.2 |
| Neither model has a result on the other 2 boards here | |||
Web designLike-for-likePublic boards, shown as published, kept out of the index50
| Benchmark | GPT-5.2 | o3 | Leader |
|---|---|---|---|
| Design Arena: websites | 1195 | 1036 | GPT-5.2 |
| Design Arena: UI components | 1190 | 1017 | GPT-5.2 |
| Design Arena: game dev | 1195 | 1034 | GPT-5.2 |
| Design Arena: data viz | 1202 | 1200 | GPT-5.2 |
| Design Arena: 3D | 1079 | Not measured | Only GPT-5.2 measured |
| Design Arena: all code categories | 1170 | 1020 | GPT-5.2 |
| 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 index302
Image generationNot comparablePublic boards, shown as published, kept out of the index0 boards, none shared
| Benchmark | GPT-5.2 | o3 | 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
o3 costs less
Repository review
50K input + 3K output tokens
o3 costs less
Agent loop
200K input + 10K output tokens
o3 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 o3 |
|---|---|---|
| Maker | OpenAI | OpenAI |
| Weights | No public weights listed | No public weights listed |
| Released | December 10, 2025 | April 16, 2025 |
| Context window | 400K | 200K |
| Input $/Mtok | $1.75 | $2.00 |
| Output $/Mtok | $14.00 | $8.00 |
| Coding index | 55.9 | Not benchmarked |
| Independent measurements behind it | 3 | 0 |
| Cost per solved task | $1.91 | $0.571 |
| Source | OpenRouter listing | OpenRouter listing |
Questions
Is GPT-5.2 better than o3 for coding?
Only GPT-5.2 has a coding score, 55.9 on the Vibe Coding Index; o3 has none, so the two cannot be ranked on coding. Pick o3 for output tokens 1.8x cheaper ($8.00 vs $14.00 per million) and a lower cost per solved task ($0.571 against $1.91).
Which is cheaper, GPT-5.2 or o3?
GPT-5.2 costs $1.75 per million input tokens and $14.00 per million output tokens; o3 costs $2.00 and $8.00. Per solved SWE-bench Verified task, GPT-5.2 costs $1.91 and o3 $0.571.
Which has the bigger context window, GPT-5.2 or o3?
GPT-5.2, with 400K tokens against 200K.
How many benchmarks is the GPT-5.2 vs o3 comparison based on?
92 benchmarks have a result for both models as of October 21, 2026. GPT-5.2's coding index rests on 3 independent benchmarks and o3's on 0. Where only one model has a result, the page says so instead of naming a winner.