Model comparison
GPT-4o mini vs Llama 2-7B
Llama 2-7B is the stronger model overall, scoring 29.1 to 25.5 on the Noometry Index.
Last verified . 21 shared benchmarks.
MetaSummary
- They share 21 benchmarks with published results for both. GPT-4o mini scores higher in 4 categories and Llama 2-7B in 4 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Llama 2-7B leads 30.7 to 10.4.
- Llama 2-7B has downloadable open weights; the other is API-only.
Side by side
| GPT-4o mini | Llama 2-7B | |
|---|---|---|
| Provider | OpenAI | Meta |
| Noometry Index | 25.5 | 29.1 |
| 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 | 29 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Llama 2-7B leads
GPT-4o mini: 22.0 (#335), Llama 2-7B: 29.2 (#307)
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| LMArena Coding | 1290 | 1002 |
| 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 2-7B: —
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| BALROG | 17.4% | — |
Reasoning Llama 2-7B leads
GPT-4o mini: 8.7 (#347), Llama 2-7B: 15.7 (#312)
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1267 | 1009 |
| Epoch Capabilities Index | 126.56 | 99.06 |
| PIQA | 88.7% | 78.8% |
| ARC-AGI-2 | 0% | — |
| SimpleBench | 10.7% | — |
| Kagi LLM Benchmark | 28.8% | — |
| LiveBench Reasoning | 32.8% | — |
| Mystery Game Puzzles | 12% | — |
| DTBench | 54.4% | — |
| LiveBench Data Analysis | 50% | — |
| LMCA | 10.4% | — |
| BIG-Bench Hard | — | 39.2% |
| HellaSwag | — | 77.2% |
| LAMBADA | — | 73.3% |
| LiveBench | 41.3% | — |
| WinoGrande | — | 69.2% |
Math Llama 2-7B leads
GPT-4o mini: 10.4 (#314), Llama 2-7B: 30.7 (#233)
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| LMArena Math | 1267 | 1042 |
| GSM8K | 91.3% | 16.7% |
| 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 2-7B leads
GPT-4o mini: 17.7 (#284), Llama 2-7B: 28.2 (#248)
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| LMArena Expert | 1235 | 1036 |
| BoolQ | 88.7% | 77.9% |
| MMLU | 81.8% | 45.8% |
| GPQA Diamond | 37.7% | — |
| SimpleQA Verified | 8.3% | — |
| MMLU-Pro | 60.3% | — |
| Confabulations | 37.2% | — |
| GPQA (HELM) | 36.8% | — |
| ARC (AI2) Challenge | — | 45.9% |
| OpenBookQA | — | 58.6% |
| TriviaQA | — | 73.7% |
Multimodal Not comparable
GPT-4o mini: 25.9 (#122), Llama 2-7B: —
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| LMArena Vision | 1066 | — |
| Video-MME | 64.8% | — |
| GeoBench | 64% | — |
| VPCT | 34% | — |
| ScienceQA | — | 43.1% |
Multilingual GPT-4o mini leads
GPT-4o mini: 42.0 (#199), Llama 2-7B: 23.8 (#293)
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| LMArena Non-English | 1266 | 973 |
| LMArena Chinese | 1265 | 973 |
| LMArena French | 1297 | 970 |
| LMArena German | 1272 | 978 |
| LMArena Russian | 1275 | 995 |
| LMArena Spanish | 1276 | 1007 |
| LMArena Japanese | 1216 | — |
| LMArena Korean | 1195 | — |
Instruction Following GPT-4o mini leads
GPT-4o mini: 61.9 (#239), Llama 2-7B: 50.8 (#298)
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| LMArena Instruction Following | 1258 | 1006 |
| LiveBench Instruction Following | 56.8% | — |
| IFEval | 78.2% | — |
Long Context GPT-4o mini leads
GPT-4o mini: 39.1 (#186), Llama 2-7B: 30.4 (#287)
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| LMArena Longer Query | 1289 | 999 |
Writing & Preference GPT-4o mini leads
GPT-4o mini: 39.5 (#248), Llama 2-7B: 28.0 (#298)
| Benchmark | GPT-4o mini | Llama 2-7B |
|---|---|---|
| LMArena Text | 1286 | 1053 |
| LMArena Creative Writing | 1268 | 1033 |
| LMArena Multi-Turn | 1285 | 1029 |
| 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 2-7B?
Llama 2-7B is the stronger model overall, scoring 29.1 to 25.5 on the Noometry Index.
Is GPT-4o mini or Llama 2-7B better for coding?
Llama 2-7B scores higher on coding benchmarks: 29.2 versus 22.0 in the Noometry coding category.
How many benchmarks do GPT-4o mini and Llama 2-7B share?
21 benchmarks have published results for both models. GPT-4o mini has 60 scored results on Noometry and Llama 2-7B has 29.