Model comparison
Gemini 2.0 Pro vs GPT-4
Gemini 2.0 Pro is the stronger model overall, scoring 39.1 to 29.1 on the Noometry Index.
Last verified . 3 shared benchmarks.
GoogleSummary
- They share 3 benchmarks with published results for both. Gemini 2.0 Pro scores higher in 6 categories and GPT-4 in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in math, where Gemini 2.0 Pro leads 39.7 to 10.8.
- The biggest single-benchmark swing is MATH Level 5: 83.5% for Gemini 2.0 Pro and 23% for GPT-4.
Side by side
| Gemini 2.0 Pro | GPT-4 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 39.1 | 29.1 |
| Released | 2025-02-05 | 2023-03-14 |
| Weights | Proprietary | Proprietary |
| Context window | — | 8K |
| Max output | — | 8K |
| Input $ / M tokens | — | $30 |
| Output $ / M tokens | — | $60 |
| Results tracked | 14 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 2.0 Pro leads
Gemini 2.0 Pro: 37.8 (#187), GPT-4: 31.6 (#283)
| Benchmark | Gemini 2.0 Pro | GPT-4 |
|---|---|---|
| Aider Polyglot | 35.6% | — |
| WeirdML | — | 12.4% |
| BigCodeBench Instruct | — | 46% |
| LiveBench Coding | 63.5% | — |
| LMArena Coding | — | 1254 |
| BigCodeBench Complete | — | 57.2% |
| HumanEval+ | — | 79.3% |
Agentic & Tool Use Not comparable
Gemini 2.0 Pro: —, GPT-4: —
| Benchmark | Gemini 2.0 Pro | GPT-4 |
|---|---|---|
| METR Time Horizons | — | 36.1% |
Reasoning Gemini 2.0 Pro leads
Gemini 2.0 Pro: 22.3 (#198), GPT-4: 17.8 (#289)
| Benchmark | Gemini 2.0 Pro | GPT-4 |
|---|---|---|
| Epoch Capabilities Index | 135.06 | 125.89 |
| Chess Puzzles | — | 4% |
| EnigmaEval | 0.7% | — |
| LiveBench Reasoning | 60.1% | — |
| LMArena Hard Prompts | — | 1241 |
| Mystery Game Puzzles | — | 12% |
| DTBench | — | 62.7% |
| LiveBench Data Analysis | 68% | — |
| LMCA | — | 17.1% |
| BIG-Bench Hard | — | 75.1% |
| ForecastBench | — | 57.8 |
| HellaSwag | — | 95.3% |
| LiveBench | 65.1% | — |
| WinoGrande | — | 87.5% |
Math Gemini 2.0 Pro leads
Gemini 2.0 Pro: 39.7 (#100), GPT-4: 10.8 (#309)
| Benchmark | Gemini 2.0 Pro | GPT-4 |
|---|---|---|
| MATH Level 5 | 83.5% | 23% |
| OTIS Mock AIME 2024-2025 | — | 1.1% |
| LiveBench Math | 71% | — |
| LMArena Math | — | 1269 |
| GSM8K | — | 92% |
Knowledge Gemini 2.0 Pro leads
Gemini 2.0 Pro: 36.5 (#167), GPT-4: 18.4 (#282)
| Benchmark | Gemini 2.0 Pro | GPT-4 |
|---|---|---|
| GPQA Diamond | 65.7% | 35.7% |
| Confabulations | 18.4% | — |
| LMArena Expert | — | 1211 |
| MMLU | — | 86.4% |
| TriviaQA | — | 84.8% |
Multilingual Not comparable
Gemini 2.0 Pro: —, GPT-4: 40.6 (#215)
| Benchmark | Gemini 2.0 Pro | GPT-4 |
|---|---|---|
| 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 Gemini 2.0 Pro leads
Gemini 2.0 Pro: 75.5 (#59), GPT-4: 65.3 (#222)
| Benchmark | Gemini 2.0 Pro | GPT-4 |
|---|---|---|
| LiveBench Instruction Following | 83.4% | — |
| LMArena Instruction Following | — | 1241 |
Long Context GPT-4 leads
Gemini 2.0 Pro: 29.2 (#292), GPT-4: 37.7 (#212)
| Benchmark | Gemini 2.0 Pro | GPT-4 |
|---|---|---|
| Fiction.LiveBench | 41.7% | — |
| LMArena Longer Query | — | 1244 |
Writing & Preference Gemini 2.0 Pro leads
Gemini 2.0 Pro: 52.7 (#165), GPT-4: 34.9 (#268)
| Benchmark | Gemini 2.0 Pro | GPT-4 |
|---|---|---|
| LMArena Text | — | 1263 |
| LMArena Creative Writing | — | 1244 |
| EQ-Bench Creative Writing | — | 752 |
| LMArena Multi-Turn | — | 1257 |
| LiveBench Language | 44.9% | — |
Frequently asked questions
Is Gemini 2.0 Pro better than GPT-4?
Gemini 2.0 Pro is the stronger model overall, scoring 39.1 to 29.1 on the Noometry Index.
Is Gemini 2.0 Pro or GPT-4 better for coding?
Gemini 2.0 Pro scores higher on coding benchmarks: 37.8 versus 31.6 in the Noometry coding category.
How many benchmarks do Gemini 2.0 Pro and GPT-4 share?
3 benchmarks have published results for both models. Gemini 2.0 Pro has 14 scored results on Noometry and GPT-4 has 38.