Loading…
Loading…
Models / multilingual-e5
1 model (~278M) · ↓ 6.3M downloads a month · from $0.08/h on GTX 1660 SUPER 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 |
|---|---|---|---|---|
| Originalintfloat/multilingual-e5-base | Text Embeddings Inference | 560 MB | 2.2 GB | $0.08/h GTX 1660 SUPER |
| base (Xenova)Xenova/multilingual-e5-base | ONNX Runtime | 280 MB | 1.4 GB | $0.08/h GTX 1660 SUPER |
| base ONNX (onnx-community)onnx-community/multilingual-e5-base-ONNX | ONNX Runtime | 280 MB | 1.4 GB | $0.08/h GTX 1660 SUPER |
| morisien embed (Singaraj)Singaraj/morisien-embed | Text Embeddings Inference | 560 MB | 2.2 GB | $0.08/h GTX 1660 SUPER |
| french me5 base (antoinelouis)antoinelouis/french-me5-base | Text Embeddings Inference | 230 MB | 1.8 GB | $0.08/h GTX 1660 SUPER |
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").