Model comparison
GPT-4 vs Llama 3.2 90B
GPT-4 is the stronger model overall, scoring 29.1 to 27.5 on the Noometry Index.
Last verified . 5 shared benchmarks.
MetaSummary
- They share 5 benchmarks with published results for both. GPT-4 scores higher in 0 categories and Llama 3.2 90B in 3 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Llama 3.2 90B leads 21.7 to 17.8.
- The biggest single-benchmark swing is MATH Level 5: 23% for GPT-4 and 39.4% for Llama 3.2 90B.
- Llama 3.2 90B has downloadable open weights; the other is API-only.
Side by side
| GPT-4 | Llama 3.2 90B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 29.1 | 27.5 |
| Released | 2023-03-14 | 2024-09-24 |
| Weights | Proprietary | Open |
| Context window | 8K | — |
| Max output | 8K | — |
| Input $ / M tokens | $30 | — |
| Output $ / M tokens | $60 | — |
| Results tracked | 38 | 9 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
GPT-4: 31.6 (#283), Llama 3.2 90B: —
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| WeirdML | 12.4% | — |
| BigCodeBench Instruct | 46% | — |
| LMArena Coding | 1254 | — |
| BigCodeBench Complete | 57.2% | — |
| HumanEval+ | 79.3% | — |
Agentic & Tool Use Not comparable
GPT-4: —, Llama 3.2 90B: 30.0 (#80)
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| BALROG | — | 27.3% |
| METR Time Horizons | 36.1% | — |
Reasoning Llama 3.2 90B leads
GPT-4: 17.8 (#289), Llama 3.2 90B: 21.7 (#217)
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| Epoch Capabilities Index | 125.89 | 125.5 |
| Chess Puzzles | 4% | — |
| EnigmaEval | — | 0.4% |
| 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 Too close to call
GPT-4: 10.8 (#309), Llama 3.2 90B: 11.1 (#308)
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 1.1% | 2.6% |
| MATH Level 5 | 23% | 39.4% |
| LMArena Math | 1269 | — |
| GSM8K | 92% | — |
Knowledge Llama 3.2 90B leads
GPT-4: 18.4 (#282), Llama 3.2 90B: 21.7 (#274)
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| GPQA Diamond | 35.7% | 41% |
| MMLU | 86.4% | 80.3% |
| LMArena Expert | 1211 | — |
| TriviaQA | 84.8% | — |
Multimodal Not comparable
GPT-4: —, Llama 3.2 90B: 25.4 (#124)
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| LMArena Vision | — | 1000 |
| GeoBench | — | 52% |
Multilingual Not comparable
GPT-4: 40.6 (#215), Llama 3.2 90B: —
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| 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), Llama 3.2 90B: —
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| LMArena Instruction Following | 1241 | — |
Long Context Not comparable
GPT-4: 37.7 (#212), Llama 3.2 90B: —
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| LMArena Longer Query | 1244 | — |
Writing & Preference Not comparable
GPT-4: 34.9 (#268), Llama 3.2 90B: —
| Benchmark | GPT-4 | Llama 3.2 90B |
|---|---|---|
| LMArena Text | 1263 | — |
| LMArena Creative Writing | 1244 | — |
| EQ-Bench Creative Writing | 752 | — |
| LMArena Multi-Turn | 1257 | — |
Frequently asked questions
Is GPT-4 better than Llama 3.2 90B?
GPT-4 is the stronger model overall, scoring 29.1 to 27.5 on the Noometry Index.
How many benchmarks do GPT-4 and Llama 3.2 90B share?
5 benchmarks have published results for both models. GPT-4 has 38 scored results on Noometry and Llama 3.2 90B has 9.