Update README.md
Browse files
README.md
CHANGED
@@ -25,6 +25,10 @@ You are Dolphin, a helpful AI assistant.<|im_end|>
|
|
25 |
- creator: Eric Hartford: [https://erichartford.com/dolphin](https://erichartford.com/dolphin)
|
26 |
- compiled by: Hrishi Olickel: [say hi on Twitter!](https://twitter.com/hrishioa)
|
27 |
|
|
|
|
|
|
|
|
|
28 |
## Description
|
29 |
|
30 |
This is a quantized version of Dolphin 2.1 🐬, one of the best finetunes of [Mistral-7b](https://huggingface.co/mistralai/Mistral-7B-v0.1) out there, ready to be used for on-browser inference over WebGPU.
|
@@ -33,8 +37,6 @@ Compiled with [mlc-llm](https://llm.mlc.ai/).
|
|
33 |
|
34 |
Very helpful direction provided by [felladrin](https://github.com/felladrin)!
|
35 |
|
36 |
-
You can use [his example](https://huggingface.co/spaces/Felladrin/Web-LLM-Mistral-7B-OpenOrca) to get quickly started with this model.
|
37 |
-
|
38 |
## Prompt template: Dolphin
|
39 |
|
40 |
Prompt format:
|
|
|
25 |
- creator: Eric Hartford: [https://erichartford.com/dolphin](https://erichartford.com/dolphin)
|
26 |
- compiled by: Hrishi Olickel: [say hi on Twitter!](https://twitter.com/hrishioa)
|
27 |
|
28 |
+
# Demo
|
29 |
+
|
30 |
+
[You can access this model over the browser here.](https://wasmai.vercel.app)
|
31 |
+
|
32 |
## Description
|
33 |
|
34 |
This is a quantized version of Dolphin 2.1 🐬, one of the best finetunes of [Mistral-7b](https://huggingface.co/mistralai/Mistral-7B-v0.1) out there, ready to be used for on-browser inference over WebGPU.
|
|
|
37 |
|
38 |
Very helpful direction provided by [felladrin](https://github.com/felladrin)!
|
39 |
|
|
|
|
|
40 |
## Prompt template: Dolphin
|
41 |
|
42 |
Prompt format:
|