# Magistral Medium vs Phi 3 Mini 4k Instruct

> Magistral Medium is the stronger model overall, scoring 35.2 to 27.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/magistral-medium-vs-phi-3-mini-4k-instruct
- Last updated: 2026-10-12
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. Magistral Medium scores higher in 7 categories and Phi 3 Mini 4k Instruct in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Magistral Medium leads 46.3 to 27.6.

## Snapshot

| | Magistral Medium | Phi 3 Mini 4k Instruct |
|---|---|---|
| Provider | Mistral AI | Microsoft |
| Noometry Index | 35.2 | 27.9 |
| Rank | 227 | 328 |
| Context | 262K | — |
| Input $/M | $2 | — |
| Output $/M | $5 | — |
| Weights | Open | Open |

## Coding

- Magistral Medium: 39.1 (#161)
- Phi 3 Mini 4k Instruct: 26.6 (#323)

| Benchmark | Magistral Medium | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Coding | 1319 | 1093 |
| SciCode | 39.2% | — |
| LiveBench Coding | — | 15.5% |
| HumanEval+ | — | 59.1% |
| MBPP+ | — | 54.2% |

## Reasoning

- Magistral Medium: 8.6 (#348)
- Phi 3 Mini 4k Instruct: 14.1 (#328)

| Benchmark | Magistral Medium | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1267 | 1072 |
| ARC-AGI-2 | 0% | — |
| Kagi LLM Benchmark | 16.2% | — |
| ARC-AGI-1 | 6.1% | — |
| CritPt | 0.3% | — |
| 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

- Magistral Medium: 35.1 (#189)
- Phi 3 Mini 4k Instruct: 26.6 (#257)

| Benchmark | Magistral Medium | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Math | 1250 | 1111 |
| LiveBench Math | — | 15.7% |

## Knowledge

- Magistral Medium: 33.5 (#202)
- Phi 3 Mini 4k Instruct: 28.5 (#246)

| Benchmark | Magistral Medium | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Expert | 1223 | 1045 |
| ARC (AI2) Challenge | — | 84.9% |
| MMLU | — | 68.8% |
| OpenBookQA | — | 88% |
| TriviaQA | — | 64% |

## Multilingual

- Magistral Medium: 39.6 (#224)
- Phi 3 Mini 4k Instruct: 26.3 (#280)

| Benchmark | Magistral Medium | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Non-English | 1232 | 1021 |
| LMArena Chinese | 1227 | 1021 |
| LMArena French | 1267 | 1076 |
| LMArena German | 1248 | 1044 |
| LMArena Japanese | 1175 | 935 |
| LMArena Korean | 1125 | 905 |
| LMArena Russian | 1224 | 1022 |
| LMArena Spanish | 1271 | 1085 |

## Instruction Following

- Magistral Medium: 66.0 (#211)
- Phi 3 Mini 4k Instruct: 47.7 (#303)

| Benchmark | Magistral Medium | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1254 | 1053 |
| LiveBench Instruction Following | — | 39.1% |

## Long Context

- Magistral Medium: 39.3 (#183)
- Phi 3 Mini 4k Instruct: 31.7 (#276)

| Benchmark | Magistral Medium | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1295 | 1044 |

## Writing & Preference

- Magistral Medium: 46.3 (#219)
- Phi 3 Mini 4k Instruct: 27.6 (#300)

| Benchmark | Magistral Medium | Phi 3 Mini 4k Instruct |
|---|---|---|
| LMArena Text | 1255 | 1073 |
| LMArena Creative Writing | 1245 | 1037 |
| LMArena Multi-Turn | 1275 | 1018 |
| LiveBench Language | — | 9.2% |

## FAQ

### Is Magistral Medium better than Phi 3 Mini 4k Instruct?

Magistral Medium is the stronger model overall, scoring 35.2 to 27.9 on the Noometry Index.

### Is Magistral Medium or Phi 3 Mini 4k Instruct better for coding?

Magistral Medium scores higher on coding benchmarks: 39.1 versus 26.6 in the Noometry coding category.

### How many benchmarks do Magistral Medium and Phi 3 Mini 4k Instruct share?

17 benchmarks have published results for both models. Magistral Medium has 22 scored results on Noometry and Phi 3 Mini 4k Instruct has 35.