Model comparison
Mistral vs Phi-2
Mistral has enough public results to be ranked (#297); Phi-2 does not yet, so treat this comparison as directional.
Last verified . 0 shared benchmarks.
Mistral AI
MicrosoftSummary
- Phi-2 has downloadable open weights; the other is API-only.
Side by side
| Mistral | Phi-2 | |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 29.4 | — |
| Released | — | 2023-12-12 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 11 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Not comparable
Mistral: 33.2 (#249), Phi-2: —
| Benchmark | Mistral | Phi-2 |
|---|---|---|
| LMArena Coding | 1162 | — |
| HumanEval+ | — | 45.1% |
| MBPP+ | — | 54.2% |
Reasoning Not comparable
Mistral: 21.3 (#199), Phi-2: —
| Benchmark | Mistral | Phi-2 |
|---|---|---|
| LMArena Hard Prompts | 1149 | — |
| Adversarial NLI | — | 42.5% |
| BIG-Bench Hard | — | 59.4% |
| Epoch Capabilities Index | — | 107.62 |
| HellaSwag | — | 53.6% |
| WinoGrande | — | 54.7% |
Math Not comparable
Mistral: 20.5 (#275), Phi-2: —
| Benchmark | Mistral | Phi-2 |
|---|---|---|
| Omni-MATH | 7.2% | — |
| LMArena Math | 1180 | — |
Knowledge Not comparable
Mistral: 16.3 (#283), Phi-2: —
| Benchmark | Mistral | Phi-2 |
|---|---|---|
| MMLU-Pro | 27.7% | — |
| GPQA (HELM) | 30.3% | — |
| LMArena Expert | 1125 | — |
| ARC (AI2) Challenge | — | 75.9% |
| MMLU | — | 58.4% |
| OpenBookQA | — | 73.6% |
| TriviaQA | — | 45.2% |
Multilingual Not comparable
Mistral: 33.0 (#249), Phi-2: —
| Benchmark | Mistral | Phi-2 |
|---|---|---|
| LMArena Non-English | 1129 | — |
| LMArena Chinese | 1109 | — |
| LMArena French | 1179 | — |
| LMArena German | 1155 | — |
| LMArena Japanese | 1012 | — |
| LMArena Korean | 1032 | — |
| LMArena Russian | 1168 | — |
| LMArena Spanish | 1143 | — |
Instruction Following Not comparable
Mistral: 52.9 (#283), Phi-2: —
| Benchmark | Mistral | Phi-2 |
|---|---|---|
| IFEval | 56.8% | — |
| LMArena Instruction Following | 1152 | — |
Long Context Not comparable
Mistral: 35.0 (#240), Phi-2: —
| Benchmark | Mistral | Phi-2 |
|---|---|---|
| LMArena Longer Query | 1154 | — |
Writing & Preference Not comparable
Mistral: 37.2 (#255), Phi-2: —
| Benchmark | Mistral | Phi-2 |
|---|---|---|
| LMArena Text | 1165 | — |
| LMArena Creative Writing | 1158 | — |
| WildBench | 66% | — |
| LMArena Multi-Turn | 1147 | — |
Frequently asked questions
Is Mistral better than Phi-2?
Mistral has enough public results to be ranked (#297); Phi-2 does not yet, so treat this comparison as directional.
How many benchmarks do Mistral and Phi-2 share?
0 benchmarks have published results for both models. Mistral has 22 scored results on Noometry and Phi-2 has 11.