Model comparison
GPT-4o mini vs Llama 2-13B
Llama 2-13B is the stronger model overall, scoring 29.6 to 25.5 on the Noometry Index.
Last verified . 24 shared benchmarks.
MetaSummary
- They share 24 benchmarks with published results for both. GPT-4o mini scores higher in 4 categories and Llama 2-13B in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Llama 2-13B leads 31.1 to 10.4.
- The biggest single-benchmark swing is DTBench: 54.4% for GPT-4o mini and 42.2% for Llama 2-13B.
- Llama 2-13B has downloadable open weights; the other is API-only.
Side by side
| GPT-4o mini | Llama 2-13B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 25.5 | 29.6 |
| Released | 2024-07-18 | 2023-07-18 |
| Weights | Proprietary | Open |
| Context window | 128K | — |
| Max output | 16K | — |
| Input $ / M tokens | $0.15 | — |
| Output $ / M tokens | $0.60 | — |
| Results tracked | 60 | 32 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 2-13B leads
GPT-4o mini: 22.0 (#335), Llama 2-13B: 30.9 (#291)
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| LMArena Coding | 1290 | 1062 |
| Aider Polyglot | 3.6% | — |
| WeirdML | 11.8% | — |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 43.1% | — |
| BigCodeBench Complete | 57.4% | — |
| HumanEval+ | 83.5% | — |
| MBPP+ | 72.2% | — |
Agentic & Tool Use Not comparable
GPT-4o mini: 27.5 (#101), Llama 2-13B: —
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| BALROG | 17.4% | — |
Reasoning Llama 2-13B leads
GPT-4o mini: 8.7 (#347), Llama 2-13B: 12.8 (#337)
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1267 | 1051 |
| DTBench | 54.4% | 42.2% |
| Epoch Capabilities Index | 126.56 | 106.17 |
| PIQA | 88.7% | 80.8% |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 10.7% | — |
| Kagi LLM Benchmark | 28.8% | — |
| LiveBench Reasoning | 32.8% | — |
| Mystery Game Puzzles | 12% | — |
| LiveBench Data Analysis | 50% | — |
| LMCA | 10.4% | — |
| BIG-Bench Hard | — | 58.2% |
| HellaSwag | — | 80.7% |
| LAMBADA | — | 76.5% |
| LiveBench | 41.3% | — |
| WinoGrande | — | 72.8% |
Math Llama 2-13B leads
GPT-4o mini: 10.4 (#314), Llama 2-13B: 31.1 (#229)
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| LMArena Math | 1267 | 1065 |
| GSM8K | 91.3% | 36.9% |
| FrontierMath (Tiers 1-3) | 0.7% | — |
| OTIS Mock AIME 2024-2025 | 6.9% | — |
| Omni-MATH | 28% | — |
| LiveBench Math | 36.3% | — |
| MATH Level 5 | 52.6% | — |
Knowledge Llama 2-13B leads
GPT-4o mini: 17.7 (#284), Llama 2-13B: 28.1 (#249)
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| LMArena Expert | 1235 | 1030 |
| BoolQ | 88.7% | 82.4% |
| MMLU | 81.8% | 55.6% |
| GPQA Diamond | 37.7% | — |
| SimpleQA Verified | 8.3% | — |
| MMLU-Pro | 60.3% | — |
| Confabulations | 37.2% | — |
| GPQA (HELM) | 36.8% | — |
| ARC (AI2) Challenge | — | 60.3% |
| OpenBookQA | — | 57% |
| TriviaQA | — | 79.6% |
Multimodal Not comparable
GPT-4o mini: 25.9 (#122), Llama 2-13B: —
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| LMArena Vision | 1066 | — |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |
| ScienceQA | — | 55.8% |
Multilingual GPT-4o mini leads
GPT-4o mini: 42.0 (#199), Llama 2-13B: 26.5 (#279)
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| LMArena Non-English | 1266 | 1024 |
| LMArena Chinese | 1265 | 1001 |
| LMArena French | 1297 | 1044 |
| LMArena German | 1272 | 1009 |
| LMArena Japanese | 1216 | 894 |
| LMArena Korean | 1195 | 953 |
| LMArena Russian | 1275 | 1055 |
| LMArena Spanish | 1276 | 1087 |
Instruction Following GPT-4o mini leads
GPT-4o mini: 61.9 (#239), Llama 2-13B: 53.3 (#287)
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| LMArena Instruction Following | 1258 | 1045 |
| LiveBench Instruction Following | 56.8% | — |
| IFEval | 78.2% | — |
Long Context GPT-4o mini leads
GPT-4o mini: 39.1 (#186), Llama 2-13B: 32.3 (#269)
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| LMArena Longer Query | 1289 | 1064 |
Writing & Preference GPT-4o mini leads
GPT-4o mini: 39.5 (#248), Llama 2-13B: 29.8 (#289)
| Benchmark | GPT-4o mini | Llama 2-13B |
|---|---|---|
| LMArena Text | 1286 | 1084 |
| LMArena Creative Writing | 1268 | 1047 |
| LMArena Multi-Turn | 1285 | 1050 |
| Short-Story Creative Writing | 67.2% | — |
| EQ-Bench Creative Writing | 873 | — |
| WildBench | 79.1% | — |
| LiveBench Language | 28.6% | — |
Frequently asked questions
Is GPT-4o mini better than Llama 2-13B?
Llama 2-13B is the stronger model overall, scoring 29.6 to 25.5 on the Noometry Index.
Is GPT-4o mini or Llama 2-13B better for coding?
Llama 2-13B scores higher on coding benchmarks: 30.9 versus 22.0 in the Noometry coding category.
How many benchmarks do GPT-4o mini and Llama 2-13B share?
24 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Llama 2-13B has 32.