# GLM-4.5 vs GLM-4.5-Air

> GLM-4.5 is the stronger model overall, scoring 41.4 to 38.7 on the Noometry Index. GLM-4.5-Air costs 2.4× 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/glm-4-5-vs-glm-4-5-air
- Last updated: 2026-10-10
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. GLM-4.5 scores higher in 7 categories and GLM-4.5-Air in 1 category; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where GLM-4.5 leads 41.1 to 34.5.
- The biggest single-benchmark swing is Kagi LLM Benchmark: 57.9% for GLM-4.5 and 43% for GLM-4.5-Air.
- GLM-4.5-Air is cheaper at $0.20 / $1.10 per million input/output tokens, against $0.60 / $2.20 for GLM-4.5.

## Snapshot

| | GLM-4.5 | GLM-4.5-Air |
|---|---|---|
| Provider | Z.ai (Zhipu) | Z.ai (Zhipu) |
| Noometry Index | 41.4 | 38.7 |
| Rank | 125 | 169 |
| Context | 131K | 131K |
| Input $/M | $0.60 | $0.20 |
| Output $/M | $2.20 | $1.10 |
| Weights | Open | Open |

## Coding

- GLM-4.5: 41.1 (#121)
- GLM-4.5-Air: 34.5 (#225)

| Benchmark | GLM-4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Coding | 1434 | 1397 |
| SWE-bench Verified (bash only) | 54.2% | — |
| GSO | — | 2.9% |
| WeirdML | 40.6% | — |
| ALE-Bench | 344.82 | — |
| AlgoTune | 1.52 | — |

## Reasoning

- GLM-4.5: 27.7 (#99)
- GLM-4.5-Air: 23.3 (#166)

| Benchmark | GLM-4.5 | GLM-4.5-Air |
|---|---|---|
| Kagi LLM Benchmark | 57.9% | 43% |
| LMArena Hard Prompts | 1430 | 1379 |
| ForecastBench | — | 59.2 |

## Math

- GLM-4.5: 36.3 (#119)
- GLM-4.5-Air: 33.8 (#166)

| Benchmark | GLM-4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Math | 1427 | 1396 |
| Omni-MATH | — | 39.1% |

## Knowledge

- GLM-4.5: 35.5 (#172)
- GLM-4.5-Air: 34.6 (#184)

| Benchmark | GLM-4.5 | GLM-4.5-Air |
|---|---|---|
| Humanity's Last Exam | 8.3% | 8.1% |
| LMArena Expert | 1433 | 1370 |
| MMLU-Pro | — | 76.2% |
| Confabulations | 11.3% | — |
| Vectara Hallucination Rate | — | 9.3% |
| GPQA (HELM) | — | 59.4% |

## Multilingual

- GLM-4.5: 53.0 (#74)
- GLM-4.5-Air: 49.4 (#130)

| Benchmark | GLM-4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Non-English | 1417 | 1366 |
| LMArena Chinese | 1465 | 1427 |
| LMArena French | 1423 | 1399 |
| LMArena German | 1406 | 1376 |
| LMArena Japanese | 1416 | 1348 |
| LMArena Korean | 1382 | 1306 |
| LMArena Russian | 1414 | 1373 |
| LMArena Spanish | 1454 | 1385 |

## Instruction Following

- GLM-4.5: 74.3 (#100)
- GLM-4.5-Air: 69.8 (#165)

| Benchmark | GLM-4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Instruction Following | 1405 | 1353 |
| IFEval | — | 81.2% |

## Long Context

- GLM-4.5: 38.2 (#196)
- GLM-4.5-Air: 41.6 (#130)

| Benchmark | GLM-4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Longer Query | 1413 | 1367 |
| Fiction.LiveBench | 58.3% | — |

## Writing & Preference

- GLM-4.5: 57.7 (#121)
- GLM-4.5-Air: 56.1 (#134)

| Benchmark | GLM-4.5 | GLM-4.5-Air |
|---|---|---|
| LMArena Text | 1430 | 1384 |
| LMArena Creative Writing | 1395 | 1343 |
| LMArena Multi-Turn | 1415 | 1371 |
| Short-Story Creative Writing | 73.4% | — |
| EQ-Bench Creative Writing | 1343 | — |
| WildBench | — | 78.9% |

## FAQ

### Is GLM-4.5 better than GLM-4.5-Air?

GLM-4.5 is the stronger model overall, scoring 41.4 to 38.7 on the Noometry Index. GLM-4.5-Air costs 2.4× less per token, which makes it the better buy when GLM-4.5's lead doesn't matter for your workload.

### Which is cheaper, GLM-4.5 or GLM-4.5-Air?

GLM-4.5-Air is cheaper. It lists at $0.20 per million input tokens and $1.10 per million output tokens; GLM-4.5 lists at $0.60 and $2.20.

### Is GLM-4.5 or GLM-4.5-Air better for coding?

GLM-4.5 scores higher on coding benchmarks: 41.1 versus 34.5 in the Noometry coding category.

### Which has the bigger context window?

Both accept 131K tokens.

### How many benchmarks do GLM-4.5 and GLM-4.5-Air share?

19 benchmarks have published results for both models. GLM-4.5 has 27 scored results on Noometry and GLM-4.5-Air has 27.