Model comparison
GPT-4.5 vs GPT-4o
GPT-4.5 is the stronger model overall, scoring 37.2 to 28.6 on the Noometry Index.
Last verified . 41 shared benchmarks.
Summary
- They share 41 benchmarks with published results for both. GPT-4.5 scores higher in 10 categories and GPT-4o in 0 categories; 10 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-4.5 leads 32.6 to 10.6.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 37.8% for GPT-4.5 and 6.4% for GPT-4o.
Side by side
| GPT-4.5 | GPT-4o | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 37.2 | 28.6 |
| Released | 2025-02-27 | 2024-05-13 |
| Weights | Proprietary | Proprietary |
| Context window | — | 128K |
| Max output | — | 16K |
| Input $ / M tokens | — | $2.50 |
| Output $ / M tokens | — | $10 |
| Results tracked | 42 | 72 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4.5 leads
GPT-4.5: 42.2 (#109), GPT-4o: 24.8 (#328)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| Aider Polyglot | 44.9% | 45.3% |
| WeirdML | 39.4% | 25.1% |
| LiveBench Coding | 75.2% | 51.4% |
| LMArena Coding | 1396 | 1297 |
| SWE-bench Verified | — | 31% |
| SWE-bench Verified (bash only) | — | 21.6% |
| GSO | — | 0% |
| BigCodeBench Instruct | — | 51.1% |
| BigCodeBench Complete | — | 61.1% |
| CadEval | — | 26% |
| HumanEval+ | — | 87.2% |
| MBPP+ | — | 72.2% |
Agentic & Tool Use GPT-4.5 leads
GPT-4.5: 27.9 (#97), GPT-4o: 21.0 (#141)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| Cybench | 17.5% | 12.5% |
| GDPval | — | 9.9% |
| TheAgentCompany | — | 8.6% |
| BALROG | — | 32.3% |
| LMArena Search | — | 1006 |
| METR Time Horizons | — | 40.8% |
Reasoning GPT-4.5 leads
GPT-4.5: 13.9 (#330), GPT-4o: 9.4 (#343)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| ARC-AGI-2 | 0.8% | 0% |
| SimpleBench | 34.5% | 17.8% |
| ARC-AGI-1 | 10.3% | 4.5% |
| EnigmaEval | 3.2% | 0.8% |
| LiveBench Reasoning | 71.1% | 55.8% |
| LMArena Hard Prompts | 1403 | 1281 |
| LiveBench Data Analysis | 64.3% | 60.9% |
| Epoch Capabilities Index | 136.74 | 128.97 |
| ForecastBench | 61.7 | 57.7 |
| LiveBench | 69% | 55.3% |
| CritPt | — | 0% |
| Chess Puzzles | — | 13% |
| DTBench | — | 64.5% |
| LMCA | — | 16.6% |
Math GPT-4.5 leads
GPT-4.5: 32.6 (#211), GPT-4o: 10.6 (#312)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 37.8% | 6.4% |
| LiveBench Math | 69.3% | 49.5% |
| LMArena Math | 1412 | 1285 |
| MATH Level 5 | 78.6% | 53.3% |
| FrontierMath (Tiers 1-3) | — | 0.4% |
| Omni-MATH | — | 29.3% |
| FrontierMath (Feb 2025 set) | — | 0.3% |
Knowledge GPT-4.5 leads
GPT-4.5: 32.5 (#211), GPT-4o: 28.8 (#242)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| GPQA Diamond | 68.7% | 49.2% |
| Humanity's Last Exam | 5.4% | 2.7% |
| Confabulations | 13.6% | 15.3% |
| LMArena Expert | 1394 | 1250 |
| SimpleQA Verified | — | 26% |
| MMLU-Pro | — | 71.3% |
| Vectara Hallucination Rate | — | 9.6% |
| GPQA (HELM) | — | 52% |
| MMLU | — | 88.1% |
Multimodal GPT-4.5 leads
GPT-4.5: 37.6 (#71), GPT-4o: 34.5 (#91)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| LMArena Vision | 1195 | 1137 |
| VPCT | 45% | 40% |
| Video-MME | — | 71.9% |
| GeoBench | — | 71% |
| ScienceQA | — | 88.5% |
Multilingual GPT-4.5 leads
GPT-4.5: 52.5 (#83), GPT-4o: 43.2 (#186)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| LMArena Non-English | 1413 | 1283 |
| LMArena Chinese | 1421 | 1277 |
| LMArena French | 1418 | 1304 |
| LMArena German | 1457 | 1282 |
| LMArena Japanese | 1416 | 1257 |
| LMArena Korean | 1392 | 1234 |
| LMArena Russian | 1419 | 1286 |
| LMArena Spanish | — | 1292 |
Instruction Following GPT-4.5 leads
GPT-4.5: 72.6 (#134), GPT-4o: 66.6 (#207)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| LiveBench Instruction Following | 72.3% | 68.6% |
| LMArena Instruction Following | 1404 | 1278 |
| IFEval | — | 81.7% |
Long Context GPT-4.5 leads
GPT-4.5: 40.4 (#155), GPT-4o: 39.4 (#179)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| Fiction.LiveBench | 63.9% | 66.7% |
| LMArena Longer Query | 1406 | 1289 |
Writing & Preference GPT-4.5 leads
GPT-4.5: 56.9 (#134), GPT-4o: 52.6 (#166)
| Benchmark | GPT-4.5 | GPT-4o |
|---|---|---|
| LMArena Text | 1417 | 1300 |
| LMArena Creative Writing | 1394 | 1292 |
| Short-Story Creative Writing | 75.6% | 81.8% |
| LMArena Multi-Turn | 1444 | 1302 |
| LiveBench Language | 61.5% | 47.6% |
| EQ-Bench Creative Writing | 1258 | — |
| WildBench | — | 82.8% |
Frequently asked questions
Is GPT-4.5 better than GPT-4o?
GPT-4.5 is the stronger model overall, scoring 37.2 to 28.6 on the Noometry Index.
Is GPT-4.5 or GPT-4o better for coding?
GPT-4.5 scores higher on coding benchmarks: 42.2 versus 24.8 in the Noometry coding category.
How many benchmarks do GPT-4.5 and GPT-4o share?
41 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-4o has 72.