# Devstral Small 2505 vs GLM-4.5

> GLM-4.5 is the stronger model overall, scoring 41.4 to 33.9 on the Noometry Index. Devstral Small 2505 costs 6.7× less per token, which makes it the better buy when GLM-4.5's lead doesn't matter for your workload.

- Canonical page: https://noometry.com/compare/devstral-small-vs-glm-4-5
- Last updated: 2026-10-10
- Shared benchmarks: 2

## Summary

- They share 2 benchmarks with published results for both. Devstral Small 2505 scores higher in 0 categories and GLM-4.5 in 2 categories; 2 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where GLM-4.5 leads 27.7 to 19.1.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 37.7% for Devstral Small 2505 and 57.9% for GLM-4.5.
- Devstral Small 2505 is cheaper at $0.10 / $0.30 per million input/output tokens, against $0.60 / $2.20 for GLM-4.5.
- GLM-4.5 accepts more context: 131K tokens versus 128K.

## Snapshot

| | Devstral Small 2505 | GLM-4.5 |
|---|---|---|
| Provider | Mistral AI | Z.ai (Zhipu) |
| Noometry Index | 33.9 | 41.4 |
| Rank | 228 | 125 |
| Context | 128K | 131K |
| Input $/M | $0.10 | $0.60 |
| Output $/M | $0.30 | $2.20 |
| Weights | Open | Open |

## Coding

- Devstral Small 2505: 38.6 (#160)
- GLM-4.5: 41.1 (#121)

| Benchmark | Devstral Small 2505 | GLM-4.5 |
|---|---|---|
| SWE-bench Verified (bash only) | 56.4% | 54.2% |
| SciCode | 28.8% | — |
| WeirdML | — | 40.6% |
| LMArena Coding | — | 1434 |
| ALE-Bench | — | 344.82 |
| AlgoTune | — | 1.52 |

## Reasoning

- Devstral Small 2505: 19.1 (#243)
- GLM-4.5: 27.7 (#99)

| Benchmark | Devstral Small 2505 | GLM-4.5 |
|---|---|---|
| Kagi LLM Benchmark | 37.7% | 57.9% |
| CritPt | 0% | — |
| LMArena Hard Prompts | — | 1430 |

## Math

- Devstral Small 2505: —
- GLM-4.5: 36.3 (#119)

| Benchmark | Devstral Small 2505 | GLM-4.5 |
|---|---|---|
| LMArena Math | — | 1427 |

## Knowledge

- Devstral Small 2505: —
- GLM-4.5: 35.5 (#172)

| Benchmark | Devstral Small 2505 | GLM-4.5 |
|---|---|---|
| Humanity's Last Exam | — | 8.3% |
| Confabulations | — | 11.3% |
| LMArena Expert | — | 1433 |

## Multilingual

- Devstral Small 2505: —
- GLM-4.5: 53.0 (#74)

| Benchmark | Devstral Small 2505 | GLM-4.5 |
|---|---|---|
| LMArena Non-English | — | 1417 |
| LMArena Chinese | — | 1465 |
| LMArena French | — | 1423 |
| LMArena German | — | 1406 |
| LMArena Japanese | — | 1416 |
| LMArena Korean | — | 1382 |
| LMArena Russian | — | 1414 |
| LMArena Spanish | — | 1454 |

## Instruction Following

- Devstral Small 2505: —
- GLM-4.5: 74.3 (#100)

| Benchmark | Devstral Small 2505 | GLM-4.5 |
|---|---|---|
| LMArena Instruction Following | — | 1405 |

## Long Context

- Devstral Small 2505: —
- GLM-4.5: 38.2 (#196)

| Benchmark | Devstral Small 2505 | GLM-4.5 |
|---|---|---|
| Fiction.LiveBench | — | 58.3% |
| LMArena Longer Query | — | 1413 |

## Writing & Preference

- Devstral Small 2505: —
- GLM-4.5: 57.7 (#121)

| Benchmark | Devstral Small 2505 | GLM-4.5 |
|---|---|---|
| LMArena Text | — | 1430 |
| LMArena Creative Writing | — | 1395 |
| Short-Story Creative Writing | — | 73.4% |
| EQ-Bench Creative Writing | — | 1343 |
| LMArena Multi-Turn | — | 1415 |

## FAQ

### Is Devstral Small 2505 better than GLM-4.5?

GLM-4.5 is the stronger model overall, scoring 41.4 to 33.9 on the Noometry Index. Devstral Small 2505 costs 6.7× less per token, which makes it the better buy when GLM-4.5's lead doesn't matter for your workload.

### Which is cheaper, Devstral Small 2505 or GLM-4.5?

Devstral Small 2505 is cheaper. It lists at $0.10 per million input tokens and $0.30 per million output tokens; GLM-4.5 lists at $0.60 and $2.20.

### Is Devstral Small 2505 or GLM-4.5 better for coding?

GLM-4.5 scores higher on coding benchmarks: 41.1 versus 38.6 in the Noometry coding category.

### Which has the bigger context window?

GLM-4.5 does, with 131K tokens against 128K.

### How many benchmarks do Devstral Small 2505 and GLM-4.5 share?

2 benchmarks have published results for both models. Devstral Small 2505 has 4 scored results on Noometry and GLM-4.5 has 27.