BriceCameron commited on
Commit
eb9e8f6
·
verified ·
1 Parent(s): cdbe323

Create app.py

Browse files
Files changed (1) hide show
  1. app.py +15 -62
app.py CHANGED
@@ -1,69 +1,22 @@
1
  import gradio as gr
2
- from huggingface_hub import InferenceClient
 
3
 
 
 
4
 
5
- def respond(
6
- message,
7
- history: list[dict[str, str]],
8
- system_message,
9
- max_tokens,
10
- temperature,
11
- top_p,
12
- hf_token: gr.OAuthToken,
13
- ):
14
- """
15
- For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
16
- """
17
- client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
18
 
19
- messages = [{"role": "system", "content": system_message}]
20
-
21
- messages.extend(history)
22
-
23
- messages.append({"role": "user", "content": message})
24
-
25
- response = ""
26
-
27
- for message in client.chat_completion(
28
- messages,
29
- max_tokens=max_tokens,
30
- stream=True,
31
- temperature=temperature,
32
- top_p=top_p,
33
- ):
34
- choices = message.choices
35
- token = ""
36
- if len(choices) and choices[0].delta.content:
37
- token = choices[0].delta.content
38
-
39
- response += token
40
- yield response
41
-
42
-
43
- """
44
- For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
45
- """
46
- chatbot = gr.ChatInterface(
47
- respond,
48
- additional_inputs=[
49
- gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
50
- gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
51
- gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
52
- gr.Slider(
53
- minimum=0.1,
54
- maximum=1.0,
55
- value=0.95,
56
- step=0.05,
57
- label="Top-p (nucleus sampling)",
58
- ),
59
- ],
60
  )
61
 
62
- with gr.Blocks() as demo:
63
- with gr.Sidebar():
64
- gr.LoginButton()
65
- chatbot.render()
66
-
67
 
68
- if __name__ == "__main__":
69
- demo.launch()
 
1
  import gradio as gr
2
+ from llama_cpp import Llama
3
+ from huggingface_hub import hf_hub_download
4
 
5
+ repo_id = "failspy/Meta-Llama-3-8B-Instruct-abliterated-v3-GGUF"
6
+ filename = "Meta-Llama-3-8B-Instruct-abliterated-v3_q4.gguf"
7
 
8
+ model_path = hf_hub_download(repo_id=repo_id, filename=filename)
 
 
 
 
 
 
 
 
 
 
 
 
9
 
10
+ llm = Llama(
11
+ model_path=model_path,
12
+ n_ctx=2048,
13
+ n_threads=2
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
  )
15
 
16
+ def respond(message, history):
17
+ prompt = f"User: {message}\nAssistant:"
18
+ response = llm(prompt, max_tokens=512, stop=["User:", "\n\n"], echo=False)
19
+ return response["choices"][0]["text"]
 
20
 
21
+ demo = gr.ChatInterface(respond)
22
+ demo.launch()