Model comparison
Mistral vs Phi 3 Mini 4k Instruct
Mistral is the stronger model overall, scoring 29.9 to 27.9 on the Noometry Index.
Last verified . 17 shared benchmarks.
Mistral AI
MicrosoftSummary
- They share 17 benchmarks with published results for both. Mistral scores higher in 6 categories and Phi 3 Mini 4k Instruct in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi 3 Mini 4k Instruct leads 28.5 to 16.6.
- Phi 3 Mini 4k Instruct has downloadable open weights; the other is API-only.
Side by side
| Mistral | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 29.9 | 27.9 |
| Released | — | 2024-04-23 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 22 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Mistral leads
Mistral: 33.8 (#250), Phi 3 Mini 4k Instruct: 26.6 (#323)
| Benchmark | Mistral | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1162 | 1093 |
| LiveBench Coding | — | 15.5% |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |
Reasoning Mistral leads
Mistral: 22.2 (#200), Phi 3 Mini 4k Instruct: 14.1 (#328)
| Benchmark | Mistral | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1149 | 1072 |
| Chess Puzzles | — | 0% |
| LiveBench Reasoning | — | 26.8% |
| LiveBench Data Analysis | — | 34.7% |
| Adversarial NLI | — | 52.8% |
| BIG-Bench Hard | — | 71.7% |
| HellaSwag | — | 76.7% |
| LiveBench | — | 22.4% |
| WinoGrande | — | 70.8% |
Math Phi 3 Mini 4k Instruct leads
Mistral: 22.3 (#278), Phi 3 Mini 4k Instruct: 26.6 (#257)
| Benchmark | Mistral | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1180 | 1111 |
| Omni-MATH | 7.2% | — |
| LiveBench Math | — | 15.7% |
Knowledge Phi 3 Mini 4k Instruct leads
Mistral: 16.6 (#288), Phi 3 Mini 4k Instruct: 28.5 (#246)
| Benchmark | Mistral | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1125 | 1045 |
| MMLU-Pro | 27.7% | — |
| GPQA (HELM) | 30.3% | — |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |
Multilingual Mistral leads
Mistral: 32.8 (#254), Phi 3 Mini 4k Instruct: 26.3 (#280)
| Benchmark | Mistral | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1129 | 1021 |
| LMArena Chinese | 1109 | 1021 |
| LMArena French | 1180 | 1076 |
| LMArena German | 1155 | 1044 |
| LMArena Japanese | 1013 | 935 |
| LMArena Korean | 1032 | 905 |
| LMArena Russian | 1168 | 1022 |
| LMArena Spanish | 1143 | 1085 |
Instruction Following Mistral leads
Mistral: 52.6 (#288), Phi 3 Mini 4k Instruct: 47.7 (#303)
| Benchmark | Mistral | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1152 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
| IFEval | 56.8% | — |
Long Context Mistral leads
Mistral: 35.0 (#245), Phi 3 Mini 4k Instruct: 31.7 (#276)
| Benchmark | Mistral | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1153 | 1044 |
Writing & Preference Mistral leads
Mistral: 37.0 (#260), Phi 3 Mini 4k Instruct: 27.6 (#300)
| Benchmark | Mistral | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1165 | 1073 |
| LMArena Creative Writing | 1158 | 1037 |
| LMArena Multi-Turn | 1147 | 1018 |
| WildBench | 66% | — |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is Mistral better than Phi 3 Mini 4k Instruct?
Mistral is the stronger model overall, scoring 29.9 to 27.9 on the Noometry Index.
Is Mistral or Phi 3 Mini 4k Instruct better for coding?
Mistral scores higher on coding benchmarks: 33.8 versus 26.6 in the Noometry coding category.
How many benchmarks do Mistral and Phi 3 Mini 4k Instruct share?
17 benchmarks have published results for both models. Mistral has 22 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.