Model comparison
Llama 2-70B vs Ministral 8B
Ministral 8B is the stronger model overall, scoring 28.2 to 24.4 on the Noometry Index.
Last verified . 15 shared benchmarks.
Meta
Mistral AISummary
- They share 15 benchmarks with published results for both. Llama 2-70B scores higher in 0 categories and Ministral 8B in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Ministral 8B leads 25.7 to 8.1.
- The biggest single-benchmark swing is MATH Level 5: 3.3% for Llama 2-70B and 14.9% for Ministral 8B.
Side by side
| Llama 2-70B | Ministral 8B | |
|---|---|---|
| Provider | Meta | Mistral AI |
| Noometry Index | 24.4 | 28.2 |
| Released | 2023-07-18 | 2024-10-01 |
| Weights | Open | Open |
| Context window | — | 262K |
| Max output | — | 262K |
| Input $ / M tokens | — | $0.15 |
| Output $ / M tokens | — | $0.15 |
| Results tracked | 35 | 17 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Ministral 8B leads
Llama 2-70B: 31.4 (#286), Ministral 8B: 35.0 (#230)
| Benchmark | Llama 2-70B | Ministral 8B |
|---|---|---|
| LMArena Coding | 1079 | 1202 |
Agentic & Tool Use Not comparable
Llama 2-70B: —, Ministral 8B: 16.4 (#148)
| Benchmark | Llama 2-70B | Ministral 8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 11.1% |
Reasoning Ministral 8B leads
Llama 2-70B: 14.4 (#325), Ministral 8B: 18.4 (#281)
| Benchmark | Llama 2-70B | Ministral 8B |
|---|---|---|
| LMArena Hard Prompts | 1073 | 1191 |
| DTBench | 41.6% | 45.7% |
| BIG-Bench Hard | 64.9% | — |
| CommonsenseQA 2.0 | 50% | — |
| Epoch Capabilities Index | 113.79 | — |
| ForecastBench | 51.4 | — |
| HellaSwag | 85.3% | — |
| LAMBADA | 78.9% | — |
| PIQA | 82.8% | — |
| WinoGrande | 80.2% | — |
Math Ministral 8B leads
Llama 2-70B: 8.1 (#326), Ministral 8B: 25.7 (#267)
| Benchmark | Llama 2-70B | Ministral 8B |
|---|---|---|
| LMArena Math | 1091 | 1188 |
| MATH Level 5 | 3.3% | 14.9% |
| OTIS Mock AIME 2024-2025 | 0% | — |
| GSM8K | 69.6% | — |
Knowledge Ministral 8B leads
Llama 2-70B: 7.4 (#310), Ministral 8B: 12.6 (#297)
| Benchmark | Llama 2-70B | Ministral 8B |
|---|---|---|
| GPQA Diamond | 26.3% | 27.1% |
| LMArena Expert | 1039 | 1170 |
| Vectara Hallucination Rate | — | 7.4% |
| ARC (AI2) Challenge | 78.3% | — |
| BoolQ | 88.6% | — |
| MMLU | 69.9% | — |
| OpenBookQA | 60.2% | — |
| TriviaQA | 87.6% | — |
Multilingual Ministral 8B leads
Llama 2-70B: 27.7 (#274), Ministral 8B: 35.1 (#247)
| Benchmark | Llama 2-70B | Ministral 8B |
|---|---|---|
| LMArena Non-English | 1045 | 1165 |
| LMArena Chinese | 995 | 1193 |
| LMArena Russian | 1083 | 1195 |
| LMArena French | 1090 | — |
| LMArena German | 1041 | — |
| LMArena Japanese | 927 | — |
| LMArena Korean | 964 | — |
| LMArena Spanish | 1143 | — |
Instruction Following Ministral 8B leads
Llama 2-70B: 54.9 (#278), Ministral 8B: 60.5 (#250)
| Benchmark | Llama 2-70B | Ministral 8B |
|---|---|---|
| LMArena Instruction Following | 1071 | 1161 |
Long Context Ministral 8B leads
Llama 2-70B: 32.3 (#270), Ministral 8B: 36.7 (#227)
| Benchmark | Llama 2-70B | Ministral 8B |
|---|---|---|
| LMArena Longer Query | 1062 | 1212 |
Writing & Preference Ministral 8B leads
Llama 2-70B: 32.3 (#279), Ministral 8B: 39.6 (#246)
| Benchmark | Llama 2-70B | Ministral 8B |
|---|---|---|
| LMArena Text | 1115 | 1191 |
| LMArena Creative Writing | 1075 | 1175 |
| LMArena Multi-Turn | 1088 | 1166 |
Frequently asked questions
Is Llama 2-70B better than Ministral 8B?
Ministral 8B is the stronger model overall, scoring 28.2 to 24.4 on the Noometry Index.
Is Llama 2-70B or Ministral 8B better for coding?
Ministral 8B scores higher on coding benchmarks: 35.0 versus 31.4 in the Noometry coding category.
How many benchmarks do Llama 2-70B and Ministral 8B share?
15 benchmarks have published results for both models. Llama 2-70B has 35 scored results on Noometry and Ministral 8B has 17.