Model comparison
GPT-4 vs GPT-5 Pro
GPT-5 Pro is the stronger model overall, scoring 46.4 to 29.1 on the Noometry Index.
Last verified . 2 shared benchmarks.
Summary
- They share 2 benchmarks with published results for both. GPT-4 scores higher in 0 categories and GPT-5 Pro in 4 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where GPT-5 Pro leads 56.7 to 18.4.
- The biggest single-benchmark swing is WeirdML: 12.4% for GPT-4 and 60.4% for GPT-5 Pro.
- GPT-4 is cheaper at $30 / $60 per million input/output tokens, against $15 / $120 for GPT-5 Pro.
- GPT-5 Pro accepts more context: 400K tokens versus 8K.
Side by side
| GPT-4 | GPT-5 Pro | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 29.1 | 46.4 |
| Released | 2023-03-14 | 2025-10-06 |
| Weights | Proprietary | Proprietary |
| Context window | 8K | 400K |
| Max output | 8K | 272K |
| Input $ / M tokens | $30 | $15 |
| Output $ / M tokens | $60 | $120 |
| Results tracked | 38 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Pro leads
GPT-4: 31.6 (#283), GPT-5 Pro: 44.0 (#80)
| Benchmark | GPT-4 | GPT-5 Pro |
|---|---|---|
| WeirdML | 12.4% | 60.4% |
| BigCodeBench Instruct | 46% | — |
| LMArena Coding | 1254 | — |
| BigCodeBench Complete | 57.2% | — |
| AlgoTune | — | 1.31 |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, GPT-5 Pro: —
| Benchmark | GPT-4 | GPT-5 Pro |
|---|---|---|
| METR Time Horizons | 36.1% | — |
Reasoning GPT-5 Pro leads
GPT-4: 17.8 (#289), GPT-5 Pro: 38.9 (#62)
| Benchmark | GPT-4 | GPT-5 Pro |
|---|---|---|
| Epoch Capabilities Index | 125.89 | 150.28 |
| ARC-AGI-2 | — | 18.3% |
| SimpleBench | — | 61.6% |
| Kagi LLM Benchmark | — | 76.8% |
| ARC-AGI-1 | — | 70.2% |
| Chess Puzzles | 4% | — |
| EnigmaEval | — | 18.8% |
| LMArena Hard Prompts | 1241 | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 62.7% | — |
| LMCA | 17.1% | — |
| BIG-Bench Hard | 75.1% | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |
Math GPT-5 Pro leads
GPT-4: 10.8 (#309), GPT-5 Pro: 48.5 (#63)
| Benchmark | GPT-4 | GPT-5 Pro |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 55.8% |
| FrontierMath Tier 4 | — | 19.5% |
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| LMArena Math | 1269 | — |
| MATH Level 5 | 23% | — |
| FrontierMath Tier 4 (v1) | — | 14.6% |
| GSM8K | 92% | — |
Knowledge GPT-5 Pro leads
GPT-4: 18.4 (#282), GPT-5 Pro: 56.7 (#42)
| Benchmark | GPT-4 | GPT-5 Pro |
|---|---|---|
| GPQA Diamond | 35.7% | — |
| Humanity's Last Exam | — | 31.6% |
| LMArena Expert | 1211 | — |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual Not comparable
GPT-4: 40.6 (#215), GPT-5 Pro: —
| Benchmark | GPT-4 | GPT-5 Pro |
|---|---|---|
| LMArena Non-English | 1246 | — |
| LMArena Chinese | 1242 | — |
| LMArena French | 1283 | — |
| LMArena German | 1251 | — |
| LMArena Japanese | 1209 | — |
| LMArena Korean | 1184 | — |
| LMArena Russian | 1251 | — |
| LMArena Spanish | 1261 | — |
Instruction Following Not comparable
GPT-4: 65.3 (#222), GPT-5 Pro: —
| Benchmark | GPT-4 | GPT-5 Pro |
|---|---|---|
| LMArena Instruction Following | 1241 | — |
Long Context Not comparable
GPT-4: 37.7 (#212), GPT-5 Pro: —
| Benchmark | GPT-4 | GPT-5 Pro |
|---|---|---|
| LMArena Longer Query | 1244 | — |
Writing & Preference Not comparable
GPT-4: 34.9 (#268), GPT-5 Pro: —
| Benchmark | GPT-4 | GPT-5 Pro |
|---|---|---|
| LMArena Text | 1263 | — |
| LMArena Creative Writing | 1244 | — |
| EQ-Bench Creative Writing | 752 | — |
| LMArena Multi-Turn | 1257 | — |
Frequently asked questions
Is GPT-4 better than GPT-5 Pro?
GPT-5 Pro is the stronger model overall, scoring 46.4 to 29.1 on the Noometry Index.
Which is cheaper, GPT-4 or GPT-5 Pro?
GPT-4 is cheaper. It lists at $30 per million input tokens and $60 per million output tokens; GPT-5 Pro lists at $15 and $120.
Is GPT-4 or GPT-5 Pro better for coding?
GPT-5 Pro scores higher on coding benchmarks: 44.0 versus 31.6 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Pro does, with 400K tokens against 8K.
How many benchmarks do GPT-4 and GPT-5 Pro share?
2 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and GPT-5 Pro has 12.