# MiniMax-M2.5 vs Phi-4 Mini

> MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 30.4 on the Noometry Index. Phi-4 Mini costs 4.0× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/minimax-m2-5-vs-phi-4-mini
- Last updated: 2026-10-10
- Shared benchmarks: 1

## Summary

- They share 1 benchmark with published results for both. MiniMax-M2.5 scores higher in 2 categories and Phi-4 Mini in 1 category; 3 gaps are clear of the uncertainty.
- The widest gap is in coding, where MiniMax-M2.5 leads 47.9 to 27.6.
- The biggest single-benchmark swing is Vectara Hallucination Rate: 9.1% for MiniMax-M2.5 and 23.5% for Phi-4 Mini.
- Phi-4 Mini is cheaper at $0.075 / $0.30 per million input/output tokens, against $0.30 / $1.20 for MiniMax-M2.5.
- MiniMax-M2.5 accepts more context: 205K tokens versus 128K.

## Snapshot

| | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| Provider | MiniMax | Microsoft |
| Noometry Index | 38.1 | 30.4 |
| Rank | 181 | 280 |
| Context | 205K | 128K |
| Input $/M | $0.30 | $0.075 |
| Output $/M | $1.20 | $0.30 |
| Weights | Open | Open |

## Coding

- MiniMax-M2.5: 47.9 (#53)
- Phi-4 Mini: 27.6 (#312)

| Benchmark | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| SWE-bench Verified (bash only) | 75.8% | — |
| LMArena WebDev | 1386 | — |
| SWE-bench Multilingual | 68.3% | — |
| SciCode | — | 10.8% |
| LMArena Coding | 1380 | — |
| ALE-Bench | 618.17 | — |

## Agentic & Tool Use

- MiniMax-M2.5: 30.2 (#70)
- Phi-4 Mini: —

| Benchmark | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| Terminal-Bench | 42.7% | — |
| Vending-Bench 2 | -23.16 | — |

## Reasoning

- MiniMax-M2.5: 17.6 (#281)
- Phi-4 Mini: 21.7 (#190)

| Benchmark | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| ARC-AGI-2 | 4.9% | — |
| Kagi LLM Benchmark | 55.2% | — |
| NYT Connections (extended) | 16.8% | — |
| ARC-AGI-1 | 63.7% | — |
| CritPt | — | 0% |
| LMArena Hard Prompts | 1372 | — |
| Epoch Capabilities Index | 146.68 | — |

## Math

- MiniMax-M2.5: 25.8 (#242)
- Phi-4 Mini: —

| Benchmark | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| ProofBench | 4% | — |
| LMArena Math | 1378 | — |

## Knowledge

- MiniMax-M2.5: 38.6 (#132)
- Phi-4 Mini: 24.8 (#261)

| Benchmark | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| Vectara Hallucination Rate | 9.1% | 23.5% |
| LMArena Expert | 1378 | — |

## Multilingual

- MiniMax-M2.5: 47.3 (#147)
- Phi-4 Mini: —

| Benchmark | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| LMArena Non-English | 1338 | — |
| LMArena Chinese | 1392 | — |
| LMArena French | 1363 | — |
| LMArena German | 1361 | — |
| LMArena Japanese | 1169 | — |
| LMArena Korean | 1232 | — |
| LMArena Russian | 1359 | — |
| LMArena Spanish | 1354 | — |

## Instruction Following

- MiniMax-M2.5: 71.7 (#143)
- Phi-4 Mini: —

| Benchmark | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| LMArena Instruction Following | 1353 | — |

## Long Context

- MiniMax-M2.5: 37.4 (#211)
- Phi-4 Mini: —

| Benchmark | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| CL-bench | 11.4% | — |
| CL-bench Life | 6.3% | — |
| LMArena Longer Query | 1367 | — |

## Writing & Preference

- MiniMax-M2.5: 54.1 (#149)
- Phi-4 Mini: —

| Benchmark | MiniMax-M2.5 | Phi-4 Mini |
|---|---|---|
| LMArena Text | 1359 | — |
| LMArena Creative Writing | 1331 | — |
| EQ-Bench Creative Writing | 1361 | — |
| LMArena Multi-Turn | 1364 | — |

## FAQ

### Is MiniMax-M2.5 better than Phi-4 Mini?

MiniMax-M2.5 is the stronger model overall, scoring 38.1 to 30.4 on the Noometry Index. Phi-4 Mini costs 4.0× less per token, which makes it the better buy when MiniMax-M2.5's lead doesn't matter for your workload.

### Which is cheaper, MiniMax-M2.5 or Phi-4 Mini?

Phi-4 Mini is cheaper. It lists at $0.075 per million input tokens and $0.30 per million output tokens; MiniMax-M2.5 lists at $0.30 and $1.20.

### Is MiniMax-M2.5 or Phi-4 Mini better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 47.9 versus 27.6 in the Noometry coding category.

### Which has the bigger context window?

MiniMax-M2.5 does, with 205K tokens against 128K.

### How many benchmarks do MiniMax-M2.5 and Phi-4 Mini share?

1 benchmark has published results for both models. MiniMax-M2.5 has 33 scored results on Noometry and Phi-4 Mini has 3.