Loading…
Loading…
Models / gte-modernbert
1 model (~149M) · ↓ 460K 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 |
|---|---|---|---|---|
| OriginalAlibaba-NLP/gte-modernbert-base | Text Embeddings Inference | 300 MB | 1.9 GB | $0.08/h GTX 1660 SUPER |
| langcache embed v1 (redis)redis/langcache-embed-v1 | Text Embeddings Inference | 300 MB | 1.9 GB | $0.08/h GTX 1660 SUPER |
| GTE ModernColBERT v1 (lightonai)lightonai/GTE-ModernColBERT-v1 | Text Embeddings Inference | 300 MB | 1.9 GB | $0.08/h GTX 1660 SUPER |
| Reason ModernColBERT (lightonai)lightonai/Reason-ModernColBERT | Text Embeddings Inference | 300 MB | 1.9 GB | $0.08/h GTX 1660 SUPER |
| Agent ModernColBERT (lightonai)lightonai/Agent-ModernColBERT | Text Embeddings Inference | 300 MB | 1.9 GB | $0.08/h GTX 1660 SUPER |
| langcache embed v2 (redis)redis/langcache-embed-v2 | Text Embeddings Inference | 300 MB | 1.9 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").