Feature Extraction
MLX
sentence-transformers
xlm-roberta
embeddings
multilingual
quantized
int8
q8
revis
text-embeddings-inference
Instructions to use mavis-ai/Multilingual-e5-large-Q8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mavis-ai/Multilingual-e5-large-Q8 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir Multilingual-e5-large-Q8 mavis-ai/Multilingual-e5-large-Q8
- sentence-transformers
How to use mavis-ai/Multilingual-e5-large-Q8 with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("mavis-ai/Multilingual-e5-large-Q8") sentences = [ "The weather is lovely today.", "It's so sunny outside!", "He drove to the stadium." ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [3, 3] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
Upload Multilingual-e5-large-Q8
Browse files
NOTICE
CHANGED
|
@@ -22,7 +22,7 @@ Embedding dimension:
|
|
| 22 |
|
| 23 |
Modification notice:
|
| 24 |
|
| 25 |
-
This repository contains a R.E.V.I.S.
|
| 26 |
Selected 2D weight tensors were converted to symmetric per-row int8
|
| 27 |
representation with per-row scale tensors.
|
| 28 |
|
|
|
|
| 22 |
|
| 23 |
Modification notice:
|
| 24 |
|
| 25 |
+
This repository contains a R.E.V.I.S. Q8 quantized distribution.
|
| 26 |
Selected 2D weight tensors were converted to symmetric per-row int8
|
| 27 |
representation with per-row scale tensors.
|
| 28 |
|
README.md
CHANGED
|
@@ -15,7 +15,7 @@ tags:
|
|
| 15 |
- revis
|
| 16 |
---
|
| 17 |
|
| 18 |
-
# mavis-ai/
|
| 19 |
|
| 20 |
This repository contains an 8-bit quantized MLX-compatible distribution of `intfloat/multilingual-e5-large`, prepared for use with **R.E.V.I.S.** as its local semantic embedding model.
|
| 21 |
|
|
@@ -35,7 +35,7 @@ For the original model card, training details, intended usage, and evaluation in
|
|
| 35 |
|
| 36 |
## Quantization
|
| 37 |
|
| 38 |
-
This package stores selected 2D weight tensors using a R.E.V.I.S.
|
| 39 |
|
| 40 |
- Quantization type: symmetric per-row int8
|
| 41 |
- Scale format: per-row scale tensor
|
|
@@ -110,11 +110,11 @@ Liang Wang, Nan Yang, Xiaolong Huang, Linjun Yang, Rangan Majumder, Furu Wei
|
|
| 110 |
Multilingual E5 Text Embeddings: A Technical Report
|
| 111 |
```
|
| 112 |
|
| 113 |
-
R.E.V.I.S.
|
| 114 |
|
| 115 |
```text
|
| 116 |
Prepared and redistributed by MAVIS / R.E.V.I.S.
|
| 117 |
-
Quantization: symmetric per-row int8
|
| 118 |
```
|
| 119 |
|
| 120 |
## Modification Notice
|
|
|
|
| 15 |
- revis
|
| 16 |
---
|
| 17 |
|
| 18 |
+
# mavis-ai/Multilingual-e5-large-Q8
|
| 19 |
|
| 20 |
This repository contains an 8-bit quantized MLX-compatible distribution of `intfloat/multilingual-e5-large`, prepared for use with **R.E.V.I.S.** as its local semantic embedding model.
|
| 21 |
|
|
|
|
| 35 |
|
| 36 |
## Quantization
|
| 37 |
|
| 38 |
+
This package stores selected 2D weight tensors using a R.E.V.I.S. Q8 format:
|
| 39 |
|
| 40 |
- Quantization type: symmetric per-row int8
|
| 41 |
- Scale format: per-row scale tensor
|
|
|
|
| 110 |
Multilingual E5 Text Embeddings: A Technical Report
|
| 111 |
```
|
| 112 |
|
| 113 |
+
R.E.V.I.S. Q8 package:
|
| 114 |
|
| 115 |
```text
|
| 116 |
Prepared and redistributed by MAVIS / R.E.V.I.S.
|
| 117 |
+
Quantization: symmetric per-row int8 Q8 package for local MLX embedding runtime
|
| 118 |
```
|
| 119 |
|
| 120 |
## Modification Notice
|