Model comparison
GPT-4o vs Llama 2-13B
GPT-4o and Llama 2-13B score almost the same on the Noometry Index (28.5 vs 29.1), so choose on price, context window or the category you care about most.
Last verified . 22 shared benchmarks.
MetaSummary
- They share 22 benchmarks with published results for both. GPT-4o scores higher in 5 categories and Llama 2-13B in 3 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-4o leads 52.8 to 30.0.
- The biggest single-benchmark swing is DTBench: 64.5% for GPT-4o and 42.2% for Llama 2-13B.
- Llama 2-13B has downloadable open weights; the other is API-only.
Side by side
| GPT-4o | Llama 2-13B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 28.5 | 29.1 |
| Released | 2024-05-13 | 2023-07-18 |
| Weights | Proprietary | Open |
| Context window | 128K | — |
| Max output | 16K | — |
| Input $ / M tokens | $2.50 | — |
| Output $ / M tokens | $10 | — |
| Results tracked | 72 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 2-13B leads
GPT-4o: 25.1 (#323), Llama 2-13B: 30.4 (#287)
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1297 | 1062 |
| SWE-bench Verified | 31% | — |
| SWE-bench Verified (bash only) | 21.6% | — |
| Aider Polyglot | 45.3% | — |
| GSO | 0% | — |
| WeirdML | 25.1% | — |
| BigCodeBench Instruct | 51.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 61.1% | — |
| CadEval | 26% | — |
| HumanEval+ | 87.2% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Not comparable
GPT-4o: 20.8 (#133), Llama 2-13B: —
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| GDPval | 9.9% | — |
| TheAgentCompany | 8.6% | — |
| Cybench | 12.5% | — |
| BALROG | 32.3% | — |
| LMArena Search | 1006 | — |
| METR Time Horizons | 40.8% | — |
Reasoning Llama 2-13B leads
GPT-4o: 9.4 (#338), Llama 2-13B: 12.3 (#333)
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| Chess Puzzles | 13% | 0% |
| LMArena Hard Prompts | 1281 | 1051 |
| DTBench | 64.5% | 42.2% |
| Epoch Capabilities Index | 128.97 | 105.8 |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 17.8% | — |
| ARC-AGI-1 | 4.5% | — |
| CritPt | 0% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 55.8% | — |
| LiveBench Data Analysis | 60.9% | — |
| LMCA | 16.6% | — |
| BIG-Bench Hard | — | 58.2% |
| ForecastBench | 57.7 | — |
| HellaSwag | — | 80.7% |
| LAMBADA | — | 76.5% |
| LiveBench | 55.3% | — |
| PIQA | — | 80.8% |
| WinoGrande | — | 72.8% |
Math Llama 2-13B leads
GPT-4o: 10.2 (#303), Llama 2-13B: 28.7 (#226)
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| LMArena Math | 1285 | 1065 |
| FrontierMath (Tiers 1-3) | 0.4% | — |
| OTIS Mock AIME 2024-2025 | 6.4% | — |
| Omni-MATH | 29.3% | — |
| LiveBench Math | 49.5% | — |
| MATH Level 5 | 53.3% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |
| GSM8K | — | 36.9% |
Knowledge GPT-4o leads
GPT-4o: 28.8 (#235), Llama 2-13B: 27.5 (#244)
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1250 | 1030 |
| MMLU | 88.1% | 55.6% |
| GPQA Diamond | 49.2% | — |
| Humanity's Last Exam | 2.7% | — |
| SimpleQA Verified | 26% | — |
| MMLU-Pro | 71.3% | — |
| Confabulations | 15.3% | — |
| Vectara Hallucination Rate | 9.6% | — |
| GPQA (HELM) | 52% | — |
| ARC (AI2) Challenge | — | 60.3% |
| BoolQ | — | 82.4% |
| OpenBookQA | — | 57% |
| TriviaQA | — | 79.6% |
Multimodal Not comparable
GPT-4o: 33.9 (#85), Llama 2-13B: —
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| ScienceQA | 88.5% | 55.8% |
| LMArena Vision | 1137 | — |
| Video-MME | 71.9% | — |
| GeoBench | 71% | — |
| VPCT | 40% | — |
Multilingual GPT-4o leads
GPT-4o: 43.4 (#181), Llama 2-13B: 26.7 (#274)
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 1283 | 1024 |
| LMArena Chinese | 1277 | 1001 |
| LMArena French | 1303 | 1043 |
| LMArena German | 1282 | 1009 |
| LMArena Japanese | 1257 | 894 |
| LMArena Korean | 1234 | 953 |
| LMArena Russian | 1286 | 1055 |
| LMArena Spanish | 1292 | 1086 |
Instruction Following GPT-4o leads
GPT-4o: 66.8 (#202), Llama 2-13B: 53.5 (#282)
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1278 | 1046 |
| LiveBench Instruction Following | 68.6% | — |
| IFEval | 81.7% | — |
Long Context GPT-4o leads
GPT-4o: 39.4 (#174), Llama 2-13B: 32.3 (#264)
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1289 | 1064 |
| Fiction.LiveBench | 66.7% | — |
Writing & Preference GPT-4o leads
GPT-4o: 52.8 (#160), Llama 2-13B: 30.0 (#284)
| Benchmark | GPT-4o | Llama 2-13B |
|---|---|---|
| LMArena Text | 1300 | 1085 |
| LMArena Creative Writing | 1291 | 1047 |
| LMArena Multi-Turn | 1302 | 1050 |
| Short-Story Creative Writing | 81.8% | — |
| WildBench | 82.8% | — |
| LiveBench Language | 47.6% | — |
Frequently asked questions
Is GPT-4o better than Llama 2-13B?
GPT-4o and Llama 2-13B score almost the same on the Noometry Index (28.5 vs 29.1), so choose on price, context window or the category you care about most.
Is GPT-4o or Llama 2-13B better for coding?
Llama 2-13B scores higher on coding benchmarks: 30.4 versus 25.1 in the Noometry coding category.
How many benchmarks do GPT-4o and Llama 2-13B share?
22 benchmarks have published results for both models. GPT-4o has 72 scored results on Noometry and Llama 2-13B has 32.