Model comparison
Llama 2-70B vs Ministral 3B
Ministral 3B is the stronger model overall, scoring 25.6 to 24.1 on the Noometry Index.
Last verified . 4 shared benchmarks.
Meta
Mistral AISummary
- They share 4 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Ministral 3B in 3 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Ministral 3B leads 24.5 to 7.5.
- The biggest single-benchmark swing is MATH Level 5: 3.3% for Llama 2-70B and 14.4% for Ministral 3B.
Side by side
| Llama 2-70B | Ministral 3B | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 24.1 | 25.6 |
| Released | 2023-07-18 | 2024-10-01 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 128K |
| Input $ / M tokens | — | $0.04 |
| Output $ / M tokens | — | $0.04 |
| Results tracked | 35 | 6 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Llama 2-70B: 30.8 (#281), Ministral 3B: —
| Benchmark | Llama 2-70B | Ministral 3B |
|---|---|---|
| LMArena Coding | 1079 | — |
Reasoning Ministral 3B leads
Llama 2-70B: 13.7 (#324), Ministral 3B: 17.7 (#278)
| Benchmark | Llama 2-70B | Ministral 3B |
|---|---|---|
| DTBench | 41.6% | 51.7% |
| Epoch Capabilities Index | 113.59 | 118.08 |
| LMArena Hard Prompts | 1073 | — |
| LMCA | — | 5.5% |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math Ministral 3B leads
Llama 2-70B: 7.5 (#320), Ministral 3B: 24.5 (#257)
| Benchmark | Llama 2-70B | Ministral 3B |
|---|---|---|
| MATH Level 5 | 3.3% | 14.4% |
| OTIS Mock AIME 2024-2025 | 0% | — |
| LMArena Math | 1091 | — |
| GSM8K | 69.6% | — |
Knowledge Ministral 3B leads
Llama 2-70B: 7.4 (#305), Ministral 3B: 10.4 (#297)
| Benchmark | Llama 2-70B | Ministral 3B |
|---|---|---|
| GPQA Diamond | 26.3% | 25.3% |
| Vectara Hallucination Rate | — | 7.3% |
| LMArena Expert | 1039 | — |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Not comparable
Llama 2-70B: 27.9 (#269), Ministral 3B: —
| Benchmark | Llama 2-70B | Ministral 3B |
|---|---|---|
| LMArena Non-English | 1045 | — |
| LMArena Chinese | 994 | — |
| LMArena French | 1089 | — |
| LMArena German | 1041 | — |
| LMArena Japanese | 927 | — |
| LMArena Korean | 964 | — |
| LMArena Russian | 1083 | — |
| LMArena Spanish | 1142 | — |
Instruction Following Not comparable
Llama 2-70B: 55.1 (#273), Ministral 3B: —
| Benchmark | Llama 2-70B | Ministral 3B |
|---|---|---|
| LMArena Instruction Following | 1071 | — |
Long Context Not comparable
Llama 2-70B: 32.2 (#265), Ministral 3B: —
| Benchmark | Llama 2-70B | Ministral 3B |
|---|---|---|
| LMArena Longer Query | 1062 | — |
Writing & Preference Not comparable
Llama 2-70B: 32.5 (#274), Ministral 3B: —
| Benchmark | Llama 2-70B | Ministral 3B |
|---|---|---|
| LMArena Text | 1115 | — |
| LMArena Creative Writing | 1075 | — |
| LMArena Multi-Turn | 1088 | — |
Frequently asked questions
Is Llama 2-70B better than Ministral 3B?
Ministral 3B is the stronger model overall, scoring 25.6 to 24.1 on the Noometry Index.
How many benchmarks do Llama 2-70B and Ministral 3B share?
4 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Ministral 3B has 6.