Model comparison
GPT-4.5 vs GPT-5 Pro
GPT-5 Pro is the stronger model overall, scoring 46.4 to 37.2 on the Noometry Index.
Last verified . 7 shared benchmarks.
Summary
- They share 7 benchmarks with published results for both. GPT-4.5 scores higher in 0 categories and GPT-5 Pro in 4 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GPT-5 Pro leads 38.9 to 13.9.
- The biggest single-benchmark swing is ARC-AGI-1: 10.3% for GPT-4.5 and 70.2% for GPT-5 Pro.
Side by side
| GPT-4.5 | GPT-5 Pro | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 37.2 | 46.4 |
| Released | 2025-02-27 | 2025-10-06 |
| Weights | Proprietary | Proprietary |
| Context window | — | 400K |
| Max output | — | 272K |
| Input $ / M tokens | — | $15 |
| Output $ / M tokens | — | $120 |
| Results tracked | 42 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Pro leads
GPT-4.5: 42.2 (#109), GPT-5 Pro: 44.0 (#80)
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| WeirdML | 39.4% | 60.4% |
| Aider Polyglot | 44.9% | — |
| LiveBench Coding | 75.2% | — |
| LMArena Coding | 1396 | — |
| AlgoTune | — | 1.31 |
Agentic & Tool Use Not comparable
GPT-4.5: 27.9 (#97), GPT-5 Pro: —
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| Cybench | 17.5% | — |
Reasoning GPT-5 Pro leads
GPT-4.5: 13.9 (#330), GPT-5 Pro: 38.9 (#62)
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| ARC-AGI-2 | 0.8% | 18.3% |
| SimpleBench | 34.5% | 61.6% |
| ARC-AGI-1 | 10.3% | 70.2% |
| EnigmaEval | 3.2% | 18.8% |
| Epoch Capabilities Index | 136.74 | 150.28 |
| Kagi LLM Benchmark | — | 76.8% |
| LiveBench Reasoning | 71.1% | — |
| LMArena Hard Prompts | 1403 | — |
| LiveBench Data Analysis | 64.3% | — |
| ForecastBench | 61.7 | — |
| LiveBench | 69% | — |
Math GPT-5 Pro leads
GPT-4.5: 32.6 (#211), GPT-5 Pro: 48.5 (#63)
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 55.8% |
| FrontierMath Tier 4 | — | 19.5% |
| OTIS Mock AIME 2024-2025 | 37.8% | — |
| LiveBench Math | 69.3% | — |
| LMArena Math | 1412 | — |
| MATH Level 5 | 78.6% | — |
| FrontierMath Tier 4 (v1) | — | 14.6% |
Knowledge GPT-5 Pro leads
GPT-4.5: 32.5 (#211), GPT-5 Pro: 56.7 (#42)
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| Humanity's Last Exam | 5.4% | 31.6% |
| GPQA Diamond | 68.7% | — |
| Confabulations | 13.6% | — |
| LMArena Expert | 1394 | — |
Multimodal Not comparable
GPT-4.5: 37.6 (#71), GPT-5 Pro: —
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| LMArena Vision | 1195 | — |
| VPCT | 45% | — |
Multilingual Not comparable
GPT-4.5: 52.5 (#83), GPT-5 Pro: —
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| LMArena Non-English | 1413 | — |
| LMArena Chinese | 1421 | — |
| LMArena French | 1418 | — |
| LMArena German | 1457 | — |
| LMArena Japanese | 1416 | — |
| LMArena Korean | 1392 | — |
| LMArena Russian | 1419 | — |
Instruction Following Not comparable
GPT-4.5: 72.6 (#134), GPT-5 Pro: —
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| LiveBench Instruction Following | 72.3% | — |
| LMArena Instruction Following | 1404 | — |
Long Context Not comparable
GPT-4.5: 40.4 (#155), GPT-5 Pro: —
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| Fiction.LiveBench | 63.9% | — |
| LMArena Longer Query | 1406 | — |
Writing & Preference Not comparable
GPT-4.5: 56.9 (#134), GPT-5 Pro: —
| Benchmark | GPT-4.5 | GPT-5 Pro |
|---|---|---|
| LMArena Text | 1417 | — |
| LMArena Creative Writing | 1394 | — |
| Short-Story Creative Writing | 75.6% | — |
| EQ-Bench Creative Writing | 1258 | — |
| LMArena Multi-Turn | 1444 | — |
| LiveBench Language | 61.5% | — |
Frequently asked questions
Is GPT-4.5 better than GPT-5 Pro?
GPT-5 Pro is the stronger model overall, scoring 46.4 to 37.2 on the Noometry Index.
Is GPT-4.5 or GPT-5 Pro better for coding?
GPT-5 Pro scores higher on coding benchmarks: 44.0 versus 42.2 in the Noometry coding category.
How many benchmarks do GPT-4.5 and GPT-5 Pro share?
7 benchmarks have published results for both models. GPT-4.5 has 42 scored results on Noometry and GPT-5 Pro has 12.