Model comparison
Mistral 7B vs Phi 3 Mini 4k Instruct
Phi 3 Mini 4k Instruct is the stronger model overall, scoring 27.4 to 22.9 on the Noometry Index.
Last verified . 27 shared benchmarks.
Mistral AI
MicrosoftSummary
- They share 27 benchmarks with published results for both. Mistral 7B scores higher in 3 categories and Phi 3 Mini 4k Instruct in 5 categories; 4 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Phi 3 Mini 4k Instruct leads 27.9 to 7.3.
Side by side
| Mistral 7B | Phi 3 Mini 4k Instruct | |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 22.9 | 27.4 |
| Released | 2023-09-27 | 2024-04-23 |
| Weights | Open | Open |
| Context window | 8K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.25 | — |
| Results tracked | 37 | 35 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
Mistral 7B: 25.9 (#321), Phi 3 Mini 4k Instruct: 26.1 (#319)
| Benchmark | Mistral 7B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1082 | 1093 |
| HumanEval+ | 36% | 59.1% |
| MBPP+ | 42.1% | 54.2% |
| BigCodeBench Instruct | 19.5% | — |
| LiveBench Coding | — | 15.5% |
| BigCodeBench Complete | 27.3% | — |
Reasoning Too close to call
Mistral 7B: 12.6 (#331), Phi 3 Mini 4k Instruct: 13.6 (#326)
| Benchmark | Mistral 7B | Phi 3 Mini 4k Instruct |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1067 | 1072 |
| Adversarial NLI | 47.1% | 52.8% |
| BIG-Bench Hard | 56.1% | 71.7% |
| HellaSwag | 81% | 76.7% |
| WinoGrande | 75.3% | 70.8% |
| LiveBench Reasoning | — | 26.8% |
| DTBench | 42.5% | — |
| LiveBench Data Analysis | — | 34.7% |
| Epoch Capabilities Index | 111.97 | — |
| LiveBench | — | 22.4% |
| PIQA | 83% | — |
Math Phi 3 Mini 4k Instruct leads
Mistral 7B: 7.6 (#319), Phi 3 Mini 4k Instruct: 24.5 (#258)
| Benchmark | Mistral 7B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1085 | 1111 |
| OTIS Mock AIME 2024-2025 | 0.3% | — |
| LiveBench Math | — | 15.7% |
| MATH Level 5 | 3.7% | — |
| GSM8K | 54.4% | — |
Knowledge Phi 3 Mini 4k Instruct leads
Mistral 7B: 7.3 (#306), Phi 3 Mini 4k Instruct: 27.9 (#241)
| Benchmark | Mistral 7B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1036 | 1045 |
| ARC (AI2) Challenge | 78.6% | 84.9% |
| MMLU | 62.5% | 68.8% |
| OpenBookQA | 79.8% | 88% |
| TriviaQA | 75.2% | 64% |
| GPQA Diamond | 15.2% | — |
| BoolQ | 87.4% | — |
Multilingual Too close to call
Mistral 7B: 26.0 (#278), Phi 3 Mini 4k Instruct: 26.5 (#275)
| Benchmark | Mistral 7B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1012 | 1021 |
| LMArena Chinese | 1009 | 1021 |
| LMArena French | 1036 | 1075 |
| LMArena German | 987 | 1044 |
| LMArena Japanese | 878 | 935 |
| LMArena Russian | 1018 | 1022 |
| LMArena Spanish | 1026 | 1084 |
| LMArena Korean | — | 905 |
Instruction Following Mistral 7B leads
Mistral 7B: 54.5 (#275), Phi 3 Mini 4k Instruct: 48.0 (#298)
| Benchmark | Mistral 7B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1060 | 1053 |
| LiveBench Instruction Following | — | 39.1% |
Long Context Too close to call
Mistral 7B: 32.2 (#266), Phi 3 Mini 4k Instruct: 31.7 (#271)
| Benchmark | Mistral 7B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1060 | 1044 |
Writing & Preference Mistral 7B leads
Mistral 7B: 30.9 (#281), Phi 3 Mini 4k Instruct: 27.7 (#295)
| Benchmark | Mistral 7B | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1090 | 1073 |
| LMArena Creative Writing | 1068 | 1037 |
| LMArena Multi-Turn | 1062 | 1018 |
| LiveBench Language | — | 9.2% |
Frequently asked questions
Is Mistral 7B better than Phi 3 Mini 4k Instruct?
Phi 3 Mini 4k Instruct is the stronger model overall, scoring 27.4 to 22.9 on the Noometry Index.
Is Mistral 7B or Phi 3 Mini 4k Instruct better for coding?
They score almost the same on coding (25.9 vs 26.1); test both on your own repository before choosing.
How many benchmarks do Mistral 7B and Phi 3 Mini 4k Instruct share?
27 benchmarks have published results for both models. Mistral 7B has 37 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.