Loading…
Loading…
Models / Vela-1.0-Encoder
GPU memory needed (smallest version of each): Vela-1.0-Encoder 307M about 2.2 GB. The cheapest runs on RTX 3060 from $0.09/h right now.
1 model (307M) · ↓ 38.4K downloads a month · from $0.09/h on RTX 3060 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 |
|---|---|---|---|---|
| Embeddingvllm-sr/Vela-1.0-Encoder-307M-Embedding | vLLM | 610 MB | 2.2 GB | $0.09/h RTX 3060 |
| Domainvllm-sr/Vela-1.0-Encoder-307M-Domain | vLLM | 620 MB | 2.2 GB | $0.09/h RTX 3060 |
| Guardvllm-sr/Vela-1.0-Encoder-307M-Guard | vLLM | 620 MB | 2.2 GB | $0.09/h RTX 3060 |
| FactCheckvllm-sr/Vela-1.0-Encoder-307M-FactCheck | vLLM | 620 MB | 2.2 GB | $0.09/h RTX 3060 |
| Feedbackvllm-sr/Vela-1.0-Encoder-307M-Feedback | vLLM | 620 MB | 2.2 GB | $0.09/h RTX 3060 |
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").