Models / granite-embedding
granite-embedding by ibm-granite
7 models (30M to 311M) · ↓ 420.7K 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.
granite-embedding 30M
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalibm-granite/granite-embedding-30m-english | Text Embeddings Inference | 60 MB | 1.6 GB | $0.08/h GTX 1660 SUPER |
| ONNX (onnx-community)onnx-community/granite-embedding-30m-english-ONNX | ONNX Runtime | 30 MB | 1 GB | $0.08/h GTX 1660 SUPER |
granite-embedding 30M
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalibm-granite/granite-embedding-30m-sparse | Text Embeddings Inference | 60 MB | 1.6 GB | $0.08/h GTX 1660 SUPER |
granite-embedding 97M
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalibm-granite/granite-embedding-97m-multilingual-r2 | Text Embeddings Inference | 190 MB | 1.8 GB | $0.08/h GTX 1660 SUPER |
| ONNX (onnx-community)onnx-community/granite-embedding-97m-multilingual-r2-ONNX | ONNX Runtime | 100 MB | 1.1 GB | $0.08/h GTX 1660 SUPER |
| IlhaEmbed (weemed)weemed/IlhaEmbed | Text Embeddings Inference | 80 MB | 1.6 GB | $0.08/h GTX 1660 SUPER |
granite-embedding 107M
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalibm-granite/granite-embedding-107m-multilingual | Text Embeddings Inference | 210 MB | 1.8 GB | $0.08/h GTX 1660 SUPER |
granite-embedding 125M
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalibm-granite/granite-embedding-125m-english | Text Embeddings Inference | 250 MB | 1.8 GB | $0.08/h GTX 1660 SUPER |
granite-embedding 278M
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalibm-granite/granite-embedding-278m-multilingual | Text Embeddings Inference | 560 MB | 2.2 GB | $0.08/h GTX 1660 SUPER |
| granite embedding 278m ita 16384 (alphaedge-ai)alphaedge-ai/granite-embedding-278m-ita-16384 | Text Embeddings Inference | 200 MB | 1.8 GB | $0.08/h GTX 1660 SUPER |
granite-embedding 311M
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|
| Originalibm-granite/granite-embedding-311m-multilingual-r2 | Text Embeddings Inference | 620 MB | 2.3 GB | $0.08/h GTX 1660 SUPER |
| lettuce eidos 768d v5 (Zeolit)Zeolit/lettuce-eidos-768d-v5 | Text Embeddings Inference | 520 MB | 2.2 GB | $0.08/h GTX 1660 SUPER |
| granite 311m italiano matryoshka (nickprock)nickprock/granite-311m-italiano-matryoshka | Text Embeddings Inference | 620 MB | 2.3 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").