# ERNIE 5.1 vs MiniMax-M2.5

> ERNIE 5.1 is the stronger model overall, scoring 43.8 to 38.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/ernie-5-1-vs-minimax-m2-5
- Last updated: 2026-10-12
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. ERNIE 5.1 scores higher in 7 categories and MiniMax-M2.5 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where ERNIE 5.1 leads 40.3 to 26.9.
- The biggest single-benchmark swing is NYT Connections (extended): 23.4% for ERNIE 5.1 and 16.8% for MiniMax-M2.5.
- MiniMax-M2.5 has downloadable open weights; the other is API-only.

## Snapshot

| | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| Provider | Baidu | MiniMax |
| Noometry Index | 43.8 | 38.3 |
| Rank | 83 | 188 |
| Context | — | 205K |
| Input $/M | — | $0.30 |
| Output $/M | — | $1.20 |
| Weights | Proprietary | Open |

## Coding

- ERNIE 5.1: 44.0 (#76)
- MiniMax-M2.5: 48.1 (#58)

| Benchmark | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Coding | 1488 | 1381 |
| SWE-bench Verified (bash only) | — | 75.8% |
| LMArena WebDev | — | 1387 |
| SWE-bench Multilingual | — | 68.3% |
| ALE-Bench | — | 618.17 |

## Agentic & Tool Use

- ERNIE 5.1: —
- MiniMax-M2.5: 30.4 (#77)

| Benchmark | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| Terminal-Bench | — | 42.7% |
| LMArena Search | 1227 | — |
| Vending-Bench 2 | — | -23.16 |

## Reasoning

- ERNIE 5.1: 21.9 (#211)
- MiniMax-M2.5: 17.5 (#292)

| Benchmark | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| NYT Connections (extended) | 23.4% | 16.8% |
| LMArena Hard Prompts | 1481 | 1372 |
| ARC-AGI-2 | — | 4.9% |
| Kagi LLM Benchmark | — | 55.2% |
| ARC-AGI-1 | — | 63.7% |
| Epoch Capabilities Index | — | 146.68 |

## Math

- ERNIE 5.1: 40.3 (#92)
- MiniMax-M2.5: 26.9 (#253)

| Benchmark | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Math | 1481 | 1378 |
| ProofBench | — | 4% |

## Knowledge

- ERNIE 5.1: 41.9 (#102)
- MiniMax-M2.5: 39.2 (#135)

| Benchmark | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Expert | 1492 | 1379 |
| Vectara Hallucination Rate | — | 9.1% |

## Multilingual

- ERNIE 5.1: 55.5 (#29)
- MiniMax-M2.5: 47.1 (#152)

| Benchmark | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1454 | 1338 |
| LMArena Chinese | 1508 | 1393 |
| LMArena French | 1488 | 1362 |
| LMArena German | 1470 | 1362 |
| LMArena Japanese | 1422 | 1171 |
| LMArena Korean | 1427 | 1232 |
| LMArena Russian | 1459 | 1358 |
| LMArena Spanish | 1473 | 1354 |

## Instruction Following

- ERNIE 5.1: 76.7 (#37)
- MiniMax-M2.5: 71.5 (#148)

| Benchmark | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1460 | 1353 |

## Long Context

- ERNIE 5.1: 44.7 (#59)
- MiniMax-M2.5: 37.5 (#216)

| Benchmark | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1462 | 1366 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |

## Writing & Preference

- ERNIE 5.1: 65.1 (#52)
- MiniMax-M2.5: 53.9 (#153)

| Benchmark | ERNIE 5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1468 | 1359 |
| LMArena Creative Writing | 1441 | 1331 |
| LMArena Multi-Turn | 1471 | 1364 |
| EQ-Bench Creative Writing | — | 1361 |

## FAQ

### Is ERNIE 5.1 better than MiniMax-M2.5?

ERNIE 5.1 is the stronger model overall, scoring 43.8 to 38.3 on the Noometry Index.

### Is ERNIE 5.1 or MiniMax-M2.5 better for coding?

MiniMax-M2.5 scores higher on coding benchmarks: 48.1 versus 44.0 in the Noometry coding category.

### How many benchmarks do ERNIE 5.1 and MiniMax-M2.5 share?

18 benchmarks have published results for both models. ERNIE 5.1 has 19 scored results on Noometry and MiniMax-M2.5 has 33.