Update README.md
Browse files
README.md
CHANGED
|
@@ -8,13 +8,18 @@ This is a repo created to keep a collection of quantized bert models in ggml for
|
|
| 8 |
### usage
|
| 9 |
You can utilize [bert.cpp](https://github.com/skeskinen/bert.cpp) as usual or use [our new api](https://rapidapi.com/nohakcoffee/api/simple-similarity) to quickly prototype on real use-case scenarios for text similarity.
|
| 10 |
|
| 11 |
-
| model | size (mb) |
|
| 12 |
-
| ---- | ---- |
|
| 13 |
-
| small | 11.2 |
|
| 14 |
-
| medium | 14.5 |
|
| 15 |
-
| large (api) | 21.3 |
|
| 16 |
|
| 17 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
|
| 19 |
### embeddings sample
|
| 20 |
|
|
|
|
| 8 |
### usage
|
| 9 |
You can utilize [bert.cpp](https://github.com/skeskinen/bert.cpp) as usual or use [our new api](https://rapidapi.com/nohakcoffee/api/simple-similarity) to quickly prototype on real use-case scenarios for text similarity.
|
| 10 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
|
| 12 |
+
| model | api | mb |
|
| 13 |
+
| ---- | ---- | ---- |
|
| 14 |
+
| l3 | pico | 11.2 |
|
| 15 |
+
| l6 | nano | 14.5 |
|
| 16 |
+
| l12 | ❌ | 21.3 |
|
| 17 |
+
| bge-small | small | 21.3 |
|
| 18 |
+
| bge-base | base | 68.8 |
|
| 19 |
+
| bge-large | ❌ | 669 |
|
| 20 |
+
|
| 21 |
+
|
| 22 |
+
We are planning to keep the list updated to the lastest open-source models.
|
| 23 |
|
| 24 |
### embeddings sample
|
| 25 |
|