Model comparison
GPT-4.5 vs GPT-5.2
GPT-5.2 is the stronger model overall, scoring 54.1 to 36.9 on the Noometry Index.
Last verified . 29 shared benchmarks.
Summary
- They share 29 benchmarks with published results for both. GPT-4.5 scores higher in 0 categories and GPT-5.2 in 10 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5.2 leads 50.1 to 14.1.
- The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 86.2% for GPT-5.2.
Side by side
| GPT-4.5 | GPT-5.2 | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 36.9 | 54.1 |
| Released | 2025-02-27 | 2025-12-11 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 128K |
| Input $ / M tokens | — | $1.75 |
| Output $ / M tokens | — | $14 |
| Results tracked | 42 | 67 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.2 leads
GPT-4.5: 41.8 (#102), GPT-5.2: 52.2 (#32)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| WeirdML | 39.4% | 72.2% |
| LMArena Coding | 1397 | 1447 |
| SWE-bench Verified | — | 73.8% |
| SWE-bench Verified (bash only) | — | 72.8% |
| Aider Polyglot | 44.9% | — |
| LMArena WebDev | — | 1415 |
| SWE-bench Multilingual | — | 66.7% |
| GSO | — | 27.4% |
| LiveBench Coding | 75.2% | — |
| ALE-Bench | — | 1,294 |
| AlgoTune | — | 2.05 |
Agentic & Tool Use GPT-5.2 leads
GPT-4.5: 27.6 (#92), GPT-5.2: 40.0 (#24)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| Terminal-Bench | — | 64.9% |
| Berkeley Function Calling Leaderboard | — | 55.9% |
| GDPval | — | 49.7% |
| Remote Labor Index | — | 2.5% |
| τ²-bench Airline | — | 83% |
| τ²-bench Banking | — | 32.2% |
| τ²-bench Retail | — | 81.6% |
| τ²-bench Telecom | — | 89.7% |
| Cybench | 17.5% | — |
| DeepResearch Bench | — | 41.1% |
| LMArena Search | — | 1207 |
| METR Time Horizons | — | 75.3% |
| Vending-Bench 2 | — | 3,591 |
Reasoning GPT-5.2 leads
GPT-4.5: 14.1 (#320), GPT-5.2: 50.1 (#33)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| ARC-AGI-2 | 0.8% | 52.9% |
| SimpleBench | 34.5% | 45.8% |
| ARC-AGI-1 | 10.3% | 86.2% |
| EnigmaEval | 3.2% | 10.4% |
| LMArena Hard Prompts | 1404 | 1445 |
| Epoch Capabilities Index | 136.74 | 153.45 |
| ForecastBench | 61.7 | 60.1 |
| Kagi LLM Benchmark | — | 73.3% |
| NYT Connections (extended) | — | 83.6% |
| Chess Puzzles | — | 49% |
| EBR-Bench | — | 23% |
| LiveBench Reasoning | 71.1% | — |
| Mystery Game Puzzles | — | 23% |
| DTBench | — | 90.9% |
| LiveBench Data Analysis | 64.3% | — |
| LMCA | — | 43.9% |
| LiveBench | 69% | — |
Math GPT-5.2 leads
GPT-4.5: 31.0 (#200), GPT-5.2: 59.8 (#32)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 96.1% |
| LMArena Math | 1412 | 1439 |
| FrontierMath (Tiers 1-3) | — | 67.4% |
| FrontierMath Tier 4 | — | 31.7% |
| MathArena Final-Answer Competitions | — | 72% |
| ProofBench | — | 15% |
| LiveBench Math | 69.3% | — |
| MATH Level 5 | 78.6% | — |
| FrontierMath (Feb 2025 set) | — | 40.7% |
| FrontierMath Tier 4 (v1) | — | 18.8% |
Knowledge GPT-5.2 leads
GPT-4.5: 32.4 (#205), GPT-5.2: 59.2 (#29)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| GPQA Diamond | 68.7% | 91.4% |
| Humanity's Last Exam | 5.4% | 27.8% |
| LMArena Expert | 1394 | 1445 |
| SimpleQA Verified | — | 37.1% |
| Confabulations | 13.6% | — |
| Vectara Hallucination Rate | — | 8.4% |
Multimodal GPT-5.2 leads
GPT-4.5: 37.0 (#68), GPT-5.2: 50.9 (#7)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| LMArena Vision | 1195 | 1268 |
| VPCT | 45% | 84% |
| Furniture Assembly | — | 38.3% |
| LMArena Document | — | 1405 |
Multilingual Too close to call
GPT-4.5: 52.7 (#78), GPT-5.2: 53.6 (#63)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| LMArena Non-English | 1413 | 1425 |
| LMArena Chinese | 1421 | 1460 |
| LMArena French | 1417 | 1456 |
| LMArena German | 1457 | 1448 |
| LMArena Japanese | 1416 | 1421 |
| LMArena Korean | 1392 | 1390 |
| LMArena Russian | 1419 | 1440 |
| LMArena Spanish | — | 1433 |
Instruction Following GPT-5.2 leads
GPT-4.5: 72.8 (#129), GPT-5.2: 74.9 (#83)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| LMArena Instruction Following | 1404 | 1417 |
| LiveBench Instruction Following | 72.3% | — |
Long Context GPT-5.2 leads
GPT-4.5: 40.4 (#150), GPT-5.2: 44.0 (#74)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| LMArena Longer Query | 1406 | 1428 |
| Fiction.LiveBench | 63.9% | — |
| CL-bench | — | 18.2% |
Writing & Preference GPT-5.2 leads
GPT-4.5: 57.1 (#129), GPT-5.2: 67.0 (#30)
| Benchmark | GPT-4.5 | GPT-5.2 |
|---|---|---|
| LMArena Text | 1417 | 1439 |
| LMArena Creative Writing | 1394 | 1402 |
| EQ-Bench Creative Writing | 1258 | 1703 |
| LMArena Multi-Turn | 1444 | 1458 |
| Short-Story Creative Writing | 75.6% | — |
| LiveBench Language | 61.5% | — |
Frequently asked questions
Is GPT-4.5 better than GPT-5.2?
GPT-5.2 is the stronger model overall, scoring 54.1 to 36.9 on the Noometry Index.
Is GPT-4.5 or GPT-5.2 better for coding?
GPT-5.2 scores higher on coding benchmarks: 52.2 versus 41.8 in the Noometry coding category.
How many benchmarks do GPT-4.5 and GPT-5.2 share?
29 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-5.2 has 67.