Model comparison
Ministral 3B vs Mistral 7B
Ministral 3B is the stronger model overall, scoring 26.2 to 23.0 on the Noometry Index.
Last verified . 4 shared benchmarks.
Mistral AISummary
- They share 4 benchmarks with published results for both. Ministral 3B scores higher in 3 categories and Mistral 7B in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Ministral 3B leads 26.6 to 8.1.
- The biggest single-benchmark swing is MATH Level 5: 14.4% for Ministral 3B and 3.7% for Mistral 7B.
- Ministral 3B is cheaper at $0.10 / $0.10 per million input/output tokens, against $0.25 / $0.25 for Mistral 7B.
- Ministral 3B accepts more context: 131K tokens versus 8K.
Side by side
| Ministral 3B | Mistral 7B | |
|---|---|---|
| Provider | Mistral AI | Mistral AI |
| Noometry Index | 26.2 | 23.0 |
| Released | 2024-10-01 | 2023-09-27 |
| Weights | Open | Open |
| Context window | 131K | 8K |
| Max output | 262K | 8K |
| Input $ / M tokens | $0.10 | $0.25 |
| Output $ / M tokens | $0.10 | $0.25 |
| Results tracked | 6 | 37 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Ministral 3B: —, Mistral 7B: 26.4 (#326)
| Benchmark | Ministral 3B | Mistral 7B |
|---|---|---|
| BigCodeBench Instruct | — | 19.5% |
| LMArena Coding | — | 1082 |
| BigCodeBench Complete | — | 27.3% |
| HumanEval+ | — | 36% |
| MBPP+ | — | 42.1% |
Reasoning Ministral 3B leads
Ministral 3B: 18.4 (#282), Mistral 7B: 13.1 (#336)
| Benchmark | Ministral 3B | Mistral 7B |
|---|---|---|
| DTBench | 51.7% | 42.5% |
| Epoch Capabilities Index | 118.1 | 112.21 |
| Chess Puzzles | — | 0% |
| LMArena Hard Prompts | — | 1067 |
| LMCA | 5.5% | — |
| Adversarial NLI | — | 47.1% |
| BIG-Bench Hard | — | 56.1% |
| HellaSwag | — | 81% |
| PIQA | — | 83% |
| WinoGrande | — | 75.3% |
Math Ministral 3B leads
Ministral 3B: 26.6 (#258), Mistral 7B: 8.1 (#325)
| Benchmark | Ministral 3B | Mistral 7B |
|---|---|---|
| MATH Level 5 | 14.4% | 3.7% |
| OTIS Mock AIME 2024-2025 | — | 0.3% |
| LMArena Math | — | 1085 |
| GSM8K | — | 54.4% |
Knowledge Ministral 3B leads
Ministral 3B: 10.4 (#302), Mistral 7B: 7.4 (#311)
| Benchmark | Ministral 3B | Mistral 7B |
|---|---|---|
| GPQA Diamond | 25.3% | 15.2% |
| Vectara Hallucination Rate | 7.3% | — |
| LMArena Expert | — | 1036 |
| ARC (AI2) Challenge | — | 78.6% |
| BoolQ | — | 87.4% |
| MMLU | — | 62.5% |
| OpenBookQA | — | 79.8% |
| TriviaQA | — | 75.2% |
Multilingual Not comparable
Ministral 3B: —, Mistral 7B: 25.8 (#283)
| Benchmark | Ministral 3B | Mistral 7B |
|---|---|---|
| LMArena Non-English | — | 1012 |
| LMArena Chinese | — | 1009 |
| LMArena French | — | 1037 |
| LMArena German | — | 987 |
| LMArena Japanese | — | 878 |
| LMArena Russian | — | 1018 |
| LMArena Spanish | — | 1026 |
Instruction Following Not comparable
Ministral 3B: —, Mistral 7B: 54.2 (#280)
| Benchmark | Ministral 3B | Mistral 7B |
|---|---|---|
| LMArena Instruction Following | — | 1060 |
Long Context Not comparable
Ministral 3B: —, Mistral 7B: 32.2 (#271)
| Benchmark | Ministral 3B | Mistral 7B |
|---|---|---|
| LMArena Longer Query | — | 1060 |
Writing & Preference Not comparable
Ministral 3B: —, Mistral 7B: 30.7 (#286)
| Benchmark | Ministral 3B | Mistral 7B |
|---|---|---|
| LMArena Text | — | 1090 |
| LMArena Creative Writing | — | 1068 |
| LMArena Multi-Turn | — | 1062 |
Frequently asked questions
Is Ministral 3B better than Mistral 7B?
Ministral 3B is the stronger model overall, scoring 26.2 to 23.0 on the Noometry Index.
Which is cheaper, Ministral 3B or Mistral 7B?
Ministral 3B is cheaper. It lists at $0.10 per million input tokens and $0.10 per million output tokens; Mistral 7B lists at $0.25 and $0.25.
Which has the bigger context window?
Ministral 3B does, with 131K tokens against 8K.
How many benchmarks do Ministral 3B and Mistral 7B share?
4 benchmarks have published results for both models. Ministral 3B has 6 scored results on Noometry and Mistral 7B has 37.