Z AI cloud models
23 models tracked via Artificial Analysis. Compare cloud performance, then find local GGUF versions in the GraySoft model catalog.
| Model | Intelligence | Speed (tok/s) |
|---|---|---|
| GLM-5.3 (Max) | 44.8 | 65.318 |
| GLM-5.3-Flash | 41.8 | 60.857 |
| GLM-5.3 (Low) | 34.3 | 66.788 |
| GLM-5.2 (Max) | 33.7 | 0 |
| GLM-5 (Reasoning) | 27.9 | 0 |
| GLM-5-Turbo | 26.6 | 0 |
| GLM-5.1 (Reasoning) | 26.1 | 0 |
| GLM-5.1 (Non-reasoning) | 24.2 | 0 |
| GLM 5V Turbo (Reasoning) | 23.5 | 0 |
| GLM-5.2 (Non-reasoning) | 22.4 | 0 |
| GLM-4.7 (Reasoning) | 22.2 | 0 |
| GLM-5 (Non-reasoning) | 21.8 | 0 |
| GLM-4.6 (Reasoning) | 18.5 | 0 |
| GLM-4.7 (Non-reasoning) | 17.4 | 0 |
| GLM-4.6 (Non-reasoning) | 14.9 | 0 |
| GLM-4.7-Flash (Reasoning) | 14.9 | 0 |
| GLM-4.5 (Reasoning) | 12.8 | 0 |
| GLM-4.6V (Reasoning) | 11.2 | 0 |
| GLM-4.5-Air | 11.1 | 0 |
| GLM-4.7-Flash (Non-reasoning) | 10.6 | 0 |
| GLM-4.6V (Non-reasoning) | 8.4 | 0 |
| GLM-4.5V (Reasoning) | 7.6 | 0 |
| GLM-4.5V (Non-reasoning) | 6.7 | 0 |
Run models locally with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.