Loading…
Loading…
Models / GLM-5.3-BF16
1 model (~753B) · ↓ 161.6K downloads a month · from $11.77/h on 8x RTX PRO 6000 right now. Each model's versions do the same job; quantized ones (AWQ, FP8, 4-bit) need less GPU memory and run on cheaper machines.
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalzai-org/GLM-5.3-BF16 | vLLM | 1507 GB | 1711.2 GB | $132.05/h 8x Nvidia B300 |
| GLM 5.3 NVFP4 (Inferact)Inferact/GLM-5.3-NVFP4 | vLLM | 465 GB | 553.6 GB | $11.77/h 8x RTX PRO 6000 |
| GLM 5.3 NVFP4 (RadixArk)RadixArk/GLM-5.3-NVFP4 | vLLM | 465 GB | 530.3 GB | $11.77/h 8x RTX PRO 6000 |
Model details come from Hugging Face and are refreshed daily; prices follow the live offers. GPU memory is GridPup's estimate: the weights, room for an 8,192-token conversation (chat models) and the runtime's own memory. Large models are split across the GPUs of one machine ("4x RTX 4090").