Model comparison
GPT-4 vs o3-pro
o3-pro is the stronger model overall, scoring 42.9 to 29.1 on the Noometry Index.
Last verified . 4 shared benchmarks.
Summary
- They share 4 benchmarks with published results for both. GPT-4 scores higher in 0 categories and o3-pro in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in long context, where o3-pro leads 72.2 to 37.7.
- The biggest single-benchmark swing is WeirdML: 12.4% for GPT-4 and 58.2% for o3-pro.
- o3-pro is cheaper at $20 / $80 per million input/output tokens, against $30 / $60 for GPT-4.
- o3-pro accepts more context: 200K tokens versus 8K.
Side by side
| GPT-4 | o3-pro | |
|---|---|---|
| Provider | OpenAI | OpenAI |
| Noometry Index | 29.1 | 42.9 |
| Released | 2023-03-14 | 2025-06-10 |
| Weights | Proprietary | Proprietary |
| Context window | 8K | 200K |
| Max output | 8K | 100K |
| Input $ / M tokens | $30 | $20 |
| Output $ / M tokens | $60 | $80 |
| Results tracked | 38 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding o3-pro leads
GPT-4: 31.6 (#283), o3-pro: 55.5 (#24)
| Benchmark | GPT-4 | o3-pro |
|---|---|---|
| WeirdML | 12.4% | 58.2% |
| Aider Polyglot | — | 84.9% |
| BigCodeBench Instruct | 46% | — |
| LMArena Coding | 1254 | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, o3-pro: —
| Benchmark | GPT-4 | o3-pro |
|---|---|---|
| METR Time Horizons | 36.1% | — |
Reasoning o3-pro leads
GPT-4: 17.8 (#289), o3-pro: 23.8 (#171)
| Benchmark | GPT-4 | o3-pro |
|---|---|---|
| DTBench | 62.7% | 86.9% |
| LMCA | 17.1% | 38.5% |
| Epoch Capabilities Index | 125.89 | 147.42 |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 72.1% |
| ARC-AGI-1 | — | 59.3% |
| Chess Puzzles | 4% | — |
| LMArena Hard Prompts | 1241 | — |
| Mystery Game Puzzles | 12% | — |
| BIG-Bench Hard | 75.1% | — |
| ForecastBench | 57.8 | — |
| HellaSwag | 95.3% | — |
| WinoGrande | 87.5% | — |
Math Not comparable
GPT-4: 10.8 (#309), o3-pro: —
| Benchmark | GPT-4 | o3-pro |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.1% | — |
| LMArena Math | 1269 | — |
| MATH Level 5 | 23% | — |
| GSM8K | 92% | — |
Knowledge o3-pro leads
GPT-4: 18.4 (#282), o3-pro: 29.5 (#238)
| Benchmark | GPT-4 | o3-pro |
|---|---|---|
| GPQA Diamond | 35.7% | — |
| Confabulations | — | 14.2% |
| Vectara Hallucination Rate | — | 23.3% |
| LMArena Expert | 1211 | — |
| MMLU | 86.4% | — |
| TriviaQA | 84.8% | — |
Multilingual Not comparable
GPT-4: 40.6 (#215), o3-pro: —
| Benchmark | GPT-4 | o3-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), o3-pro: —
| Benchmark | GPT-4 | o3-pro |
|---|---|---|
| LMArena Instruction Following | 1241 | — |
Long Context o3-pro leads
GPT-4: 37.7 (#212), o3-pro: 72.2 (#1)
| Benchmark | GPT-4 | o3-pro |
|---|---|---|
| Fiction.LiveBench | — | 97.2% |
| LMArena Longer Query | 1244 | — |
Writing & Preference o3-pro leads
GPT-4: 34.9 (#268), o3-pro: 57.1 (#133)
| Benchmark | GPT-4 | o3-pro |
|---|---|---|
| LMArena Text | 1263 | — |
| LMArena Creative Writing | 1244 | — |
| Short-Story Creative Writing | — | 84.4% |
| EQ-Bench Creative Writing | 752 | — |
| LMArena Multi-Turn | 1257 | — |
Frequently asked questions
Is GPT-4 better than o3-pro?
o3-pro is the stronger model overall, scoring 42.9 to 29.1 on the Noometry Index.
Which is cheaper, GPT-4 or o3-pro?
o3-pro is cheaper. It lists at $20 per million input tokens and $80 per million output tokens; GPT-4 lists at $30 and $60.
Is GPT-4 or o3-pro better for coding?
o3-pro scores higher on coding benchmarks: 55.5 versus 31.6 in the Noometry coding category.
Which has the bigger context window?
o3-pro does, with 200K tokens against 8K.
How many benchmarks do GPT-4 and o3-pro share?
4 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and o3-pro has 12.