# MiniMax-M2.5 vs Phi 3 Medium 4k Instruct

> MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 32.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/minimax-m2-5-vs-phi-3-medium-4k-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 17

## Summary

- They share 17 benchmarks with published results for both. MiniMax-M2.5 scores higher in 6 categories and Phi 3 Medium 4k Instruct in 2 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where MiniMax-M2.5 leads 54.1 to 34.3.

## Snapshot

| | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| Provider | MiniMax | Microsoft |
| Noometry Index | 38.1 | 32.4 |
| Rank | 181 | 246 |
| Context | 205K | — |
| Input $/M | $0.30 | — |
| Output $/M | $1.20 | — |
| Weights | Open | Open |

## Coding

- MiniMax-M2.5: 47.9 (#53)
- Phi 3 Medium 4k Instruct: 32.3 (#263)

| Benchmark | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Coding | 1380 | 1130 |
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1386 | — |
| SWE-bench Multilingual | 68.3% | — |
| ALE-Bench | 618.17 | — |

## Agentic & Tool Use

- MiniMax-M2.5: 30.2 (#70)
- Phi 3 Medium 4k Instruct: —

| Benchmark | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Vending-Bench 2 | -23.16 | — |

## Reasoning

- MiniMax-M2.5: 17.6 (#281)
- Phi 3 Medium 4k Instruct: 20.8 (#214)

| Benchmark | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Hard Prompts | 1372 | 1127 |
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| Epoch Capabilities Index | 146.68 | — |

## Math

- MiniMax-M2.5: 25.8 (#242)
- Phi 3 Medium 4k Instruct: 30.9 (#202)

| Benchmark | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Math | 1378 | 1173 |
| ProofBench | 4% | — |

## Knowledge

- MiniMax-M2.5: 38.6 (#132)
- Phi 3 Medium 4k Instruct: 29.6 (#226)

| Benchmark | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Expert | 1378 | 1108 |
| Vectara Hallucination Rate | 9.1% | — |

## Multilingual

- MiniMax-M2.5: 47.3 (#147)
- Phi 3 Medium 4k Instruct: 30.8 (#258)

| Benchmark | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Non-English | 1338 | 1094 |
| LMArena Chinese | 1392 | 1108 |
| LMArena French | 1363 | 1109 |
| LMArena German | 1361 | 1101 |
| LMArena Japanese | 1169 | 1042 |
| LMArena Korean | 1232 | 954 |
| LMArena Russian | 1359 | 1145 |
| LMArena Spanish | 1354 | 1095 |

## Instruction Following

- MiniMax-M2.5: 71.7 (#143)
- Phi 3 Medium 4k Instruct: 57.8 (#263)

| Benchmark | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Instruction Following | 1353 | 1114 |

## Long Context

- MiniMax-M2.5: 37.4 (#211)
- Phi 3 Medium 4k Instruct: 33.9 (#250)

| Benchmark | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Longer Query | 1367 | 1120 |
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |

## Writing & Preference

- MiniMax-M2.5: 54.1 (#149)
- Phi 3 Medium 4k Instruct: 34.3 (#267)

| Benchmark | MiniMax-M2.5 | Phi 3 Medium 4k Instruct |
|---|---|---|
| LMArena Text | 1359 | 1138 |
| LMArena Creative Writing | 1331 | 1106 |
| LMArena Multi-Turn | 1364 | 1088 |
| EQ-Bench Creative Writing | 1361 | — |

## FAQ

### Is MiniMax-M2.5 better than Phi 3 Medium 4k Instruct?

MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 32.4 on the Noometry Index.

### Is MiniMax-M2.5 or Phi 3 Medium 4k Instruct better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 32.3 in the Noometry coding category.

### How many benchmarks do MiniMax-M2.5 and Phi 3 Medium 4k Instruct share?

17 benchmarks have published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Phi 3 Medium 4k Instruct has 17.