# Hy4 preview vs MiniMax-M2.5

> Hy4 preview is the stronger model overall, scoring 45.3 to 38.3 on the Noometry Index. MiniMax-M2.5 costs 2.4× less per token, which makes it the better buy when Hy4 preview's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/hy4-preview-vs-minimax-m2-5
- Last updated: 2026-10-11
- Shared benchmarks: 3

## Summary

- They share 3 benchmarks with published results for both. Hy4 preview scores higher in 3 categories and MiniMax-M2.5 in 0 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in math, where Hy4 preview leads 55.7 to 26.9.
- The biggest single-benchmark swing is ProofBench: 75% for Hy4 preview and 4% for MiniMax-M2.5.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $0.83 / $2.50 for Hy4 preview.
- Hy4 preview accepts more context: 1.05M tokens versus 205K.

## Snapshot

| | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| Provider | Tencent | MiniMax |
| Noometry Index | 45.3 | 38.3 |
| Rank | 73 | 188 |
| Context | 1.05M | 205K |
| Input $/M | $0.83 | $0.30 |
| Output $/M | $2.50 | $1.20 |
| Weights | Open | Open |

## Coding

- Hy4 preview: 51.6 (#38)
- MiniMax-M2.5: 48.1 (#58)

| Benchmark | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| LMArena WebDev | 1632 | 1387 |
| SWE-bench Verified (bash only) | — | 75.8% |
| SWE-bench Multilingual | — | 68.3% |
| LMArena Coding | — | 1381 |
| ALE-Bench | — | 618.17 |

## Agentic & Tool Use

- Hy4 preview: —
- MiniMax-M2.5: 30.4 (#77)

| Benchmark | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| Vending-Bench 2 | — | -23.16 |

## Reasoning

- Hy4 preview: 31.9 (#79)
- MiniMax-M2.5: 17.5 (#292)

| Benchmark | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| NYT Connections (extended) | 68.2% | 16.8% |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 55.2% |
| ARC-AGI-1 | — | 63.7% |
| LMArena Hard Prompts | — | 1372 |
| Epoch Capabilities Index | — | 146.68 |

## Math

- Hy4 preview: 55.7 (#42)
- MiniMax-M2.5: 26.9 (#253)

| Benchmark | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| ProofBench | 75% | 4% |
| LMArena Math | — | 1378 |

## Knowledge

- Hy4 preview: —
- MiniMax-M2.5: 39.2 (#135)

| Benchmark | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| Vectara Hallucination Rate | — | 9.1% |
| LMArena Expert | — | 1379 |

## Multilingual

- Hy4 preview: —
- MiniMax-M2.5: 47.1 (#152)

| Benchmark | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | — | 1338 |
| LMArena Chinese | — | 1393 |
| LMArena French | — | 1362 |
| LMArena German | — | 1362 |
| LMArena Japanese | — | 1171 |
| LMArena Korean | — | 1232 |
| LMArena Russian | — | 1358 |
| LMArena Spanish | — | 1354 |

## Instruction Following

- Hy4 preview: —
- MiniMax-M2.5: 71.5 (#148)

| Benchmark | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | — | 1353 |

## Long Context

- Hy4 preview: —
- MiniMax-M2.5: 37.5 (#216)

| Benchmark | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |
| LMArena Longer Query | — | 1366 |

## Writing & Preference

- Hy4 preview: —
- MiniMax-M2.5: 53.9 (#153)

| Benchmark | Hy4 preview | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | — | 1359 |
| LMArena Creative Writing | — | 1331 |
| EQ-Bench Creative Writing | — | 1361 |
| LMArena Multi-Turn | — | 1364 |

## FAQ

### Is Hy4 preview better than MiniMax-M2.5?

Hy4 preview is the stronger model overall, scoring 45.3 to 38.3 on the Noometry Index. MiniMax-M2.5 costs 2.4× less per token, which makes it the better buy when Hy4 preview's lead doesn't matter for your workload.

### Which is cheaper, Hy4 preview or MiniMax-M2.5?

MiniMax-M2.5 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Hy4 preview lists at $0.83 and $2.50.

### Is Hy4 preview or MiniMax-M2.5 better for coding?

Hy4 preview scores higher on coding benchmarks: 51.6 versus 48.1 in the Noometry coding category.

### Which has the bigger context window?

Hy4 preview does, with 1.05M tokens against 205K.

### How many benchmarks do Hy4 preview and MiniMax-M2.5 share?

3 benchmarks have published results for both models. Hy4 preview has 3 scored results on Noometry and MiniMax-M2.5 has 33.