Loading…
Loading…
Models / Mistral-Medium-3.5
GPU memory needed (smallest version of each): Mistral-Medium-3.5 128B about 111.3 GB. The cheapest runs on 8x RTX 5060 Ti from $1.13/h right now.
1 model (128B) · ↓ 6.8K downloads a month · from $1.13/h on 8x RTX 5060 Ti 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 |
|---|---|---|---|---|
| NVFP4 (nvidia)nvidia/Mistral-Medium-3.5-128B-NVFP4 | vLLM | 95 GB | 111.3 GB | $1.13/h 8x RTX 5060 Ti |
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").