eleeter commited on
Commit
3b2207c
·
verified ·
1 Parent(s): 65389b8

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +43 -57
app.py CHANGED
@@ -1,69 +1,55 @@
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
+ print("Loading model... please wait.")
6
+ model_path = hf_hub_download(
7
+ repo_id="mradermacher/Qwen2.5-Coder-32B-Instruct-GGUF",
8
+ filename="Qwen2.5-Coder-32B-Instruct.IQ2_M.gguf"
9
+ )
10
 
11
+ llm = Llama(
12
+ model_path=model_path,
13
+ n_ctx=2048,
14
+ n_threads=2
15
+ )
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
16
 
17
+ def respond(message, history, system_message, max_tokens, temperature, top_p):
18
+ formatted_messages = [{"role": "system", "content": system_message}]
19
+ for user_msg, assistant_msg in history:
20
+ formatted_messages.append({"role": "user", "content": user_msg})
21
+ formatted_messages.append({"role": "assistant", "content": assistant_msg})
22
+ formatted_messages.append({"role": "user", "content": message})
23
+
24
+ response_text = ""
25
+
26
+ stream = llm.create_chat_completion(
27
+ messages=formatted_messages,
28
  max_tokens=max_tokens,
 
29
  temperature=temperature,
30
  top_p=top_p,
31
+ stream=True
32
+ )
 
 
 
 
 
 
33
 
34
+ for chunk in stream:
35
+ if "content" in chunk["choices"][0]["delta"]:
36
+ token = chunk["choices"][0]["delta"]["content"]
37
+ response_text += token
38
+ yield response_text
39
 
40
+ # --- Gradio UI Setup ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
41
  with gr.Blocks() as demo:
42
+ gr.Markdown("Qwen2.5-Coder-32B (2-bit)")
43
+
44
+ chatbot = gr.ChatInterface(
45
+ respond,
46
+ additional_inputs=[
47
+ gr.Textbox(value="You are an Elite Minecraft Modder.", label="System message"),
48
+ gr.Slider(minimum=1, maximum=1024, value=512, step=1, label="Max new tokens"),
49
+ gr.Slider(minimum=0.1, maximum=1.0, value=0.7, step=0.1, label="Temperature"),
50
+ gr.Slider(minimum=0.1, maximum=1.0, value=0.9, step=0.05, label="Top-p"),
51
+ ],
52
+ )
53
 
54
  if __name__ == "__main__":
55
+ demo.launch()