Model comparison
GPT-4o vs GPT-5.2
GPT-5.2 is the stronger model overall, scoring 54.1 to 28.6 on the Noometry Index.
Last verified . 42 shared benchmarks.
Summary
- They share 42 benchmarks with published results for both. GPT-4o scores higher in 0 categories and GPT-5.2 in 10 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5.2 leads 60.0 to 10.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 6.4% for GPT-4o and 96.1% for GPT-5.2.
- GPT-4o is cheaper at $2.50 / $10 per million input/output tokens, against $1.75 / $14 for GPT-5.2.
- GPT-5.2 accepts more context: 400K tokens versus 128K.
Side by side
| GPT-4o | GPT-5.2 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 28.6 | 54.1 |
| Released | 2024-05-13 | 2025-12-11 |
| Weights | Proprietary | Proprietary |
| Context window | 128K | 400K |
| Max output | 16K | 128K |
| Input $ / M tokens | $2.50 | $1.75 |
| Output $ / M tokens | $10 | $14 |
| Results tracked | 72 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 leads
GPT-4o: 24.8 (#328), GPT-5.2: 51.6 (#37)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| SWE-bench Verified | 31% | 73.8% |
| SWE-bench Verified (bash only) | 21.6% | 72.8% |
| GSO | 0% | 27.4% |
| WeirdML | 25.1% | 72.2% |
| LMArena Coding | 1297 | 1447 |
| Aider Polyglot | 45.3% | — |
| LMArena WebDev | — | 1416 |
| SWE-bench Multilingual | — | 66.7% |
| BigCodeBench Instruct | 51.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 61.1% | — |
| CadEval | 26% | — |
| ALE-Bench | — | 1,294 |
| AlgoTune | — | 2.05 |
| HumanEval+ | 87.2% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use GPT-5.2 leads
GPT-4o: 21.0 (#141), GPT-5.2: 40.2 (#24)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| GDPval | 9.9% | 49.7% |
| LMArena Search | 1006 | 1207 |
| METR Time Horizons | 40.8% | 75.3% |
| Terminal-Bench | — | 64.9% |
| Berkeley Function Calling Leaderboard | — | 55.9% |
| Remote Labor Index | — | 2.5% |
| TheAgentCompany | 8.6% | — |
| τ²-bench Airline | — | 83% |
| τ²-bench Banking | — | 32.2% |
| τ²-bench Retail | — | 81.6% |
| τ²-bench Telecom | — | 89.7% |
| Cybench | 12.5% | — |
| DeepResearch Bench | — | 41.1% |
| BALROG | 32.3% | — |
| Vending-Bench 2 | — | 3,591 |
Reasoning GPT-5.2 leads
GPT-4o: 9.4 (#343), GPT-5.2: 50.2 (#35)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| ARC-AGI-2 | 0% | 52.9% |
| SimpleBench | 17.8% | 45.8% |
| ARC-AGI-1 | 4.5% | 86.2% |
| Chess Puzzles | 13% | 49% |
| EnigmaEval | 0.8% | 10.4% |
| LMArena Hard Prompts | 1281 | 1445 |
| DTBench | 64.5% | 90.9% |
| LMCA | 16.6% | 43.9% |
| Epoch Capabilities Index | 128.97 | 153.45 |
| ForecastBench | 57.7 | 60.1 |
| Kagi LLM Benchmark | — | 73.3% |
| NYT Connections (extended) | — | 83.6% |
| CritPt | 0% | — |
| EBR-Bench | — | 23% |
| LiveBench Reasoning | 55.8% | — |
| Mystery Game Puzzles | — | 23% |
| LiveBench Data Analysis | 60.9% | — |
| LiveBench | 55.3% | — |
Math GPT-5.2 leads
GPT-4o: 10.6 (#312), GPT-5.2: 60.0 (#38)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 0.4% | 67.4% |
| OTIS Mock AIME 2024-2025 | 6.4% | 96.1% |
| LMArena Math | 1285 | 1440 |
| FrontierMath (Feb 2025 set) | 0.3% | 40.7% |
| FrontierMath Tier 4 | — | 31.7% |
| MathArena Final-Answer Competitions | — | 72% |
| ProofBench | — | 15% |
| Omni-MATH | 29.3% | — |
| LiveBench Math | 49.5% | — |
| MATH Level 5 | 53.3% | — |
| FrontierMath Tier 4 (v1) | — | 18.8% |
Knowledge GPT-5.2 leads
GPT-4o: 28.8 (#242), GPT-5.2: 59.3 (#32)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| GPQA Diamond | 49.2% | 91.4% |
| Humanity's Last Exam | 2.7% | 27.8% |
| SimpleQA Verified | 26% | 37.1% |
| Vectara Hallucination Rate | 9.6% | 8.4% |
| LMArena Expert | 1250 | 1445 |
| MMLU-Pro | 71.3% | — |
| Confabulations | 15.3% | — |
| GPQA (HELM) | 52% | — |
| MMLU | 88.1% | — |
Multimodal GPT-5.2 leads
GPT-4o: 34.5 (#91), GPT-5.2: 51.3 (#7)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| LMArena Vision | 1137 | 1268 |
| VPCT | 40% | 84% |
| Video-MME | 71.9% | — |
| GeoBench | 71% | — |
| Furniture Assembly | — | 38.3% |
| LMArena Document | — | 1405 |
| ScienceQA | 88.5% | — |
Multilingual GPT-5.2 leads
GPT-4o: 43.2 (#186), GPT-5.2: 53.4 (#67)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| LMArena Non-English | 1283 | 1425 |
| LMArena Chinese | 1277 | 1460 |
| LMArena French | 1304 | 1455 |
| LMArena German | 1282 | 1448 |
| LMArena Japanese | 1257 | 1420 |
| LMArena Korean | 1234 | 1392 |
| LMArena Russian | 1286 | 1440 |
| LMArena Spanish | 1292 | 1433 |
Instruction Following GPT-5.2 leads
GPT-4o: 66.6 (#207), GPT-5.2: 74.7 (#89)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| LMArena Instruction Following | 1278 | 1417 |
| LiveBench Instruction Following | 68.6% | — |
| IFEval | 81.7% | — |
Long Context GPT-5.2 leads
GPT-4o: 39.4 (#179), GPT-5.2: 44.0 (#78)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| LMArena Longer Query | 1289 | 1428 |
| Fiction.LiveBench | 66.7% | — |
| CL-bench | — | 18.2% |
Writing & Preference GPT-5.2 leads
GPT-4o: 52.6 (#166), GPT-5.2: 66.8 (#32)
| Benchmark | GPT-4o | GPT-5.2 |
|---|---|---|
| LMArena Text | 1300 | 1439 |
| LMArena Creative Writing | 1292 | 1401 |
| LMArena Multi-Turn | 1302 | 1458 |
| Short-Story Creative Writing | 81.8% | — |
| EQ-Bench Creative Writing | — | 1703 |
| WildBench | 82.8% | — |
| LiveBench Language | 47.6% | — |
Frequently asked questions
Is GPT-4o better than GPT-5.2?
GPT-5.2 is the stronger model overall, scoring 54.1 to 28.6 on the Noometry Index.
Which is cheaper, GPT-4o or GPT-5.2?
GPT-4o is cheaper. It lists at $2.50 per million input tokens and $10 per million output tokens; GPT-5.2 lists at $1.75 and $14.
Is GPT-4o or GPT-5.2 better for coding?
GPT-5.2 scores higher on coding benchmarks: 51.6 versus 24.8 in the Noometry coding category.
Which has the bigger context window?
GPT-5.2 does, with 400K tokens against 128K.
How many benchmarks do GPT-4o and GPT-5.2 share?
42 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and GPT-5.2 has 67.