Loading…
Loading…
Models / GLM-5.3
1 model (~753B) · ↓ 116.1K downloads a month · from $6.17/h on 4x 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 |
|---|---|---|---|---|
| GLM 5.3 CYBERSECURITY FP8 (dealignai)dealignai/GLM-5.3-CYBERSECURITY-FP8 | vLLM | 756 GB | 876.7 GB | $44.03/h 8x Nvidia H200 |
| GLM 5.3 UNCENSORED EXL3 3.0bpw (Infatoshi)Infatoshi/GLM-5.3-UNCENSORED-EXL3-3.0bpw | vLLM | 293 GB | 362.3 GB | $6.17/h 4x RTX PRO 6000 |
| GLM 5.3 UNCENSORED FP8 (dealignai)dealignai/GLM-5.3-UNCENSORED-FP8 | vLLM | 756 GB | 876.7 GB | $44.03/h 8x Nvidia H200 |
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").