Loading…
Loading…
Models / DeciLM
GPU memory needed (smallest version of each): DeciLM 6B about 17.3 GB, DeciLM 7B about 22.6 GB. The cheapest runs on RTX 4000 Ada from $0.22/h right now.
3 models (6B to 7B) · ↓ 1.7K downloads a month · from $0.22/h on RTX 4000 Ada 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 |
|---|---|---|---|---|
| OriginalDeci/DeciLM-6b | vLLM | 11 GB | 17.3 GB | $0.22/h RTX 4000 Ada |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalDeci/DeciLM-6b-instruct | vLLM | 11 GB | 17.3 GB | $0.22/h RTX 4000 Ada |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalDeci/DeciLM-7B | vLLM | 14 GB | 22.6 GB | $0.23/h RTX A5000 |
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").