# GLM-4.5-Air vs Hy3

> Hy3 is the stronger model overall, scoring 43.6 to 38.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/glm-4-5-air-vs-hy3
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. GLM-4.5-Air scores higher in 0 categories and Hy3 in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in coding, where Hy3 leads 46.6 to 34.5.
- Hy3 is cheaper at $0.13 / $0.53 per million input/output tokens, against $0.20 / $1.10 for GLM-4.5-Air.
- Hy3 accepts more context: 262K tokens versus 131K.

## Snapshot

| | GLM-4.5-Air | Hy3 |
|---|---|---|
| Provider | Z.ai (Zhipu) | Tencent |
| Noometry Index | 38.7 | 43.6 |
| Rank | 169 | 79 |
| Context | 131K | 262K |
| Input $/M | $0.20 | $0.13 |
| Output $/M | $1.10 | $0.53 |
| Weights | Open | Open |

## Coding

- GLM-4.5-Air: 34.5 (#225)
- Hy3: 46.6 (#60)

| Benchmark | GLM-4.5-Air | Hy3 |
|---|---|---|
| LMArena Coding | 1397 | 1465 |
| LMArena WebDev | — | 1509 |
| GSO | 2.9% | — |

## Reasoning

- GLM-4.5-Air: 23.3 (#166)
- Hy3: 25.0 (#134)

| Benchmark | GLM-4.5-Air | Hy3 |
|---|---|---|
| LMArena Hard Prompts | 1379 | 1449 |
| Kagi LLM Benchmark | 43% | — |
| NYT Connections (extended) | — | 41.2% |
| ForecastBench | 59.2 | — |

## Math

- GLM-4.5-Air: 33.8 (#166)
- Hy3: 37.4 (#98)

| Benchmark | GLM-4.5-Air | Hy3 |
|---|---|---|
| LMArena Math | 1396 | 1475 |
| Omni-MATH | 39.1% | — |

## Knowledge

- GLM-4.5-Air: 34.6 (#184)
- Hy3: 40.2 (#108)

| Benchmark | GLM-4.5-Air | Hy3 |
|---|---|---|
| LMArena Expert | 1370 | 1461 |
| Humanity's Last Exam | 8.1% | — |
| MMLU-Pro | 76.2% | — |
| Vectara Hallucination Rate | 9.3% | — |
| GPQA (HELM) | 59.4% | — |

## Multilingual

- GLM-4.5-Air: 49.4 (#130)
- Hy3: 53.7 (#60)

| Benchmark | GLM-4.5-Air | Hy3 |
|---|---|---|
| LMArena Non-English | 1366 | 1427 |
| LMArena Chinese | 1427 | 1496 |
| LMArena French | 1399 | 1461 |
| LMArena Korean | 1306 | 1390 |
| LMArena Russian | 1373 | 1435 |
| LMArena Spanish | 1385 | 1448 |
| LMArena German | 1376 | — |
| LMArena Japanese | 1348 | — |

## Instruction Following

- GLM-4.5-Air: 69.8 (#165)
- Hy3: 75.3 (#68)

| Benchmark | GLM-4.5-Air | Hy3 |
|---|---|---|
| LMArena Instruction Following | 1353 | 1425 |
| IFEval | 81.2% | — |

## Long Context

- GLM-4.5-Air: 41.6 (#130)
- Hy3: 44.1 (#70)

| Benchmark | GLM-4.5-Air | Hy3 |
|---|---|---|
| LMArena Longer Query | 1367 | 1443 |

## Writing & Preference

- GLM-4.5-Air: 56.1 (#134)
- Hy3: 62.5 (#75)

| Benchmark | GLM-4.5-Air | Hy3 |
|---|---|---|
| LMArena Text | 1384 | 1442 |
| LMArena Creative Writing | 1343 | 1399 |
| LMArena Multi-Turn | 1371 | 1436 |
| WildBench | 78.9% | — |

## FAQ

### Is GLM-4.5-Air better than Hy3?

Hy3 is the stronger model overall, scoring 43.6 to 38.7 on the Noometry Index.

### Which is cheaper, GLM-4.5-Air or Hy3?

Hy3 is cheaper. It lists at $0.13 per million input tokens and $0.53 per million output tokens; GLM-4.5-Air lists at $0.20 and $1.10.

### Is GLM-4.5-Air or Hy3 better for coding?

Hy3 scores higher on coding benchmarks: 46.6 versus 34.5 in the Noometry coding category.

### Which has the bigger context window?

Hy3 does, with 262K tokens against 131K.

### How many benchmarks do GLM-4.5-Air and Hy3 share?

15 benchmarks have published results for both models. GLM-4.5-Air has 27 scored results on Noometry and Hy3 has 17.