Loading…
Loading…
Models / sarvam-m
GPU memory needed (smallest version of each): sarvam-m ~23.6B about 56.1 GB. The cheapest runs on 4x RTX 5060 Ti from $0.73/h right now.
1 model (~23.6B) · ↓ 3.2K downloads a month · from $0.73/h on 4x 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 |
|---|---|---|---|---|
| Originalsarvamai/sarvam-m | vLLM | 47 GB | 56.1 GB | $0.73/h 4x RTX 5060 Ti |
| onnx (Prince-1)Prince-1/sarvam-m-onnx | ONNX Runtime GenAI | 47 GB | 57.1 GB | $0.93/h Nvidia A800 |
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").