# GLM-5.1 vs MiniMax-M2.5

> GLM-5.1 is the stronger model overall, scoring 47.6 to 38.1 on the Noometry Index. MiniMax-M2.5 costs 4.1× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/glm-5-1-vs-minimax-m2-5
- Last updated: 2026-10-10
- Shared benchmarks: 24

## Summary

- They share 24 benchmarks with published results for both. GLM-5.1 scores higher in 8 categories and MiniMax-M2.5 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.1 leads 49.0 to 25.8.
- The biggest single-benchmark swing is NYT Connections (extended): 77.7% for GLM-5.1 and 16.8% for MiniMax-M2.5.
- MiniMax-M2.5 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.40 / $4.40 for GLM-5.1.
- MiniMax-M2.5 accepts more context: 205K tokens versus 200K.

## Snapshot

| | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| Provider | Z.ai (Zhipu) | MiniMax |
| Noometry Index | 47.6 | 38.1 |
| Rank | 51 | 181 |
| Context | 200K | 205K |
| Input $/M | $1.40 | $0.30 |
| Output $/M | $4.40 | $1.20 |
| Weights | Open | Open |

## Coding

- GLM-5.1: 48.5 (#52)
- MiniMax-M2.5: 47.9 (#53)

| Benchmark | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena WebDev | 1509 | 1386 |
| LMArena Coding | 1486 | 1380 |
| ALE-Bench | 887.1 | 618.17 |
| SWE-bench Verified | 74.2% | — |
| SWE-bench Verified (bash only) | — | 75.8% |
| SWE-bench Multilingual | — | 68.3% |
| SciCode | 43.8% | — |
| WeirdML | 57.1% | — |

## Agentic & Tool Use

- GLM-5.1: 24.8 (#106)
- MiniMax-M2.5: 30.2 (#70)

| Benchmark | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| Vending-Bench 2 | 5,634 | -23.16 |
| Terminal-Bench | — | 42.7% |
| APEX-Agents | 40.9% | — |
| ExploitBench | 18.1% | — |
| GBAEval | 0% | — |

## Reasoning

- GLM-5.1: 38.2 (#57)
- MiniMax-M2.5: 17.6 (#281)

| Benchmark | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| NYT Connections (extended) | 77.7% | 16.8% |
| LMArena Hard Prompts | 1472 | 1372 |
| Epoch Capabilities Index | 149.86 | 146.68 |
| ARC-AGI-2 | — | 4.9% |
| SimpleBench | 55.1% | — |
| Kagi LLM Benchmark | — | 55.2% |
| ARC-AGI-1 | — | 63.7% |
| CritPt | 4.6% | — |
| Chess Puzzles | 19% | — |
| Thematic Generalization | 69.8% | — |

## Math

- GLM-5.1: 49.0 (#48)
- MiniMax-M2.5: 25.8 (#242)

| Benchmark | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| ProofBench | 22.2% | 4% |
| LMArena Math | 1472 | 1378 |
| FrontierMath (Tiers 1-3) | 36.8% | — |
| MathArena Final-Answer Competitions | 67.1% | — |
| OTIS Mock AIME 2024-2025 | 93.3% | — |
| FrontierMath (Feb 2025 set) | 33.4% | — |
| FrontierMath Tier 4 (v1) | 12.5% | — |

## Knowledge

- GLM-5.1: 55.0 (#45)
- MiniMax-M2.5: 38.6 (#132)

| Benchmark | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Expert | 1477 | 1378 |
| GPQA Diamond | 89.9% | — |
| SimpleQA Verified | 34% | — |
| Vectara Hallucination Rate | — | 9.1% |

## Multilingual

- GLM-5.1: 55.2 (#34)
- MiniMax-M2.5: 47.3 (#147)

| Benchmark | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Non-English | 1448 | 1338 |
| LMArena Chinese | 1516 | 1392 |
| LMArena French | 1478 | 1363 |
| LMArena German | 1466 | 1361 |
| LMArena Japanese | 1436 | 1169 |
| LMArena Korean | 1423 | 1232 |
| LMArena Russian | 1453 | 1359 |
| LMArena Spanish | 1467 | 1354 |

## Instruction Following

- GLM-5.1: 76.5 (#41)
- MiniMax-M2.5: 71.7 (#143)

| Benchmark | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Instruction Following | 1451 | 1353 |

## Long Context

- GLM-5.1: 44.9 (#51)
- MiniMax-M2.5: 37.4 (#211)

| Benchmark | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Longer Query | 1467 | 1367 |
| CL-bench | — | 11.4% |
| CL-bench Life | — | 6.3% |

## Writing & Preference

- GLM-5.1: 67.2 (#29)
- MiniMax-M2.5: 54.1 (#149)

| Benchmark | GLM-5.1 | MiniMax-M2.5 |
|---|---|---|
| LMArena Text | 1462 | 1359 |
| LMArena Creative Writing | 1454 | 1331 |
| EQ-Bench Creative Writing | 1592 | 1361 |
| LMArena Multi-Turn | 1473 | 1364 |

## FAQ

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

GLM-5.1 is the stronger model overall, scoring 47.6 to 38.1 on the Noometry Index. MiniMax-M2.5 costs 4.1× less per token, which makes it the better buy when GLM-5.1's lead doesn't matter for your workload.

### Which is cheaper, GLM-5.1 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; GLM-5.1 lists at $1.40 and $4.40.

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

They score almost the same on coding (48.5 vs 47.9); test both on your own repository before choosing.

### Which has the bigger context window?

MiniMax-M2.5 does, with 205K tokens against 200K.

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

24 benchmarks have published results for both models. GLM-5.1 has 41 scored results on Noometry and MiniMax-M2.5 has 33.