Model comparison
GPT-4 Turbo vs Llama 13b
GPT-4 Turbo is the stronger model overall, scoring 30.5 to 24.4 on the Noometry Index.
Last verified . 10 shared benchmarks.
MetaSummary
- They share 10 benchmarks with published results for both. GPT-4 Turbo scores higher in 5 categories and Llama 13b in 1 category; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4 Turbo leads 47.7 to 13.8.
- Llama 13b has downloadable open weights; the other is API-only.
Side by side
| GPT-4 Turbo | Llama 13b | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 30.5 | 24.4 |
| Released | 2023-11-06 | 2023-02-24 |
| Weights | Proprietary | Open |
| Context window | 128K | — |
| Max output | 4K | — |
| Input $ / M tokens | $10 | — |
| Output $ / M tokens | $30 | — |
| Results tracked | 36 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-4 Turbo leads
GPT-4 Turbo: 33.8 (#249), Llama 13b: 21.4 (#337)
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| LMArena Coding | 1268 | 683 |
| WeirdML | 18% | — |
| BigCodeBench Instruct | 48.2% | — |
| BigCodeBench Complete | 58.2% | — |
| HumanEval+ | 86.6% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
GPT-4 Turbo: —, Llama 13b: —
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| METR Time Horizons | 36.7% | — |
Reasoning GPT-4 Turbo leads
GPT-4 Turbo: 15.3 (#317), Llama 13b: 14.0 (#329)
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| LMArena Hard Prompts | 1251 | 728 |
| Epoch Capabilities Index | 127.25 | 100.58 |
| SimpleBench | 25.1% | — |
| Chess Puzzles | 6% | — |
| DTBench | 61.6% | — |
| LMCA | 9.8% | — |
| BIG-Bench Hard | — | 37.9% |
| ForecastBench | 59.4 | — |
| HellaSwag | — | 79.2% |
| LAMBADA | — | 75.2% |
| PIQA | — | 80.1% |
| WinoGrande | — | 73% |
Math Llama 13b leads
GPT-4 Turbo: 9.0 (#322), Llama 13b: 26.7 (#256)
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| LMArena Math | 1272 | 838 |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| OTIS Mock AIME 2024-2025 | 6.7% | — |
| MATH Level 5 | 46.7% | — |
| GSM8K | — | 20.6% |
Knowledge Not comparable
GPT-4 Turbo: 24.3 (#268), Llama 13b: —
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| MMLU | 81.3% | 47.7% |
| GPQA Diamond | 46.6% | — |
| Confabulations | 28.4% | — |
| LMArena Expert | 1223 | — |
| ARC (AI2) Challenge | — | 52.7% |
| BoolQ | — | 78.7% |
| OpenBookQA | — | 56.4% |
| TriviaQA | — | 77.9% |
Multimodal Not comparable
GPT-4 Turbo: 30.6 (#110), Llama 13b: —
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| LMArena Vision | 1090 | — |
| ScienceQA | — | 43.3% |
Multilingual GPT-4 Turbo leads
GPT-4 Turbo: 40.5 (#216), Llama 13b: 16.6 (#297)
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| LMArena Non-English | 1245 | 819 |
| LMArena Chinese | 1242 | — |
| LMArena French | 1276 | — |
| LMArena German | 1259 | — |
| LMArena Japanese | 1194 | — |
| LMArena Korean | 1187 | — |
| LMArena Russian | 1259 | — |
| LMArena Spanish | 1260 | — |
Instruction Following GPT-4 Turbo leads
GPT-4 Turbo: 65.8 (#216), Llama 13b: 36.7 (#305)
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| LMArena Instruction Following | 1249 | 781 |
Long Context Not comparable
GPT-4 Turbo: 38.0 (#206), Llama 13b: —
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| LMArena Longer Query | 1254 | — |
Writing & Preference GPT-4 Turbo leads
GPT-4 Turbo: 47.7 (#206), Llama 13b: 13.8 (#312)
| Benchmark | GPT-4 Turbo | Llama 13b |
|---|---|---|
| LMArena Text | 1272 | 834 |
| LMArena Creative Writing | 1269 | 794 |
| LMArena Multi-Turn | 1267 | 753 |
Frequently asked questions
Is GPT-4 Turbo better than Llama 13b?
GPT-4 Turbo is the stronger model overall, scoring 30.5 to 24.4 on the Noometry Index.
Is GPT-4 Turbo or Llama 13b better for coding?
GPT-4 Turbo scores higher on coding benchmarks: 33.8 versus 21.4 in the Noometry coding category.
How many benchmarks do GPT-4 Turbo and Llama 13b share?
10 benchmarks have published results for both models. GPT-4 Turbo has 36 scored results on Noometry and Llama 13b has 21.