Update app.py

app.py

@@ -5,7 +5,7 @@ from typing import Iterator
 import gradio as gr
 import spaces
 import torch
-from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
 MAX_MAX_NEW_TOKENS = 2048
 DEFAULT_MAX_NEW_TOKENS = 1024
@@ -13,26 +13,21 @@ MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 
 DESCRIPTION = """\
 # L-MChat
-
 This Space demonstrates [L-MChat](https://huggingface.co/collections/Artples/l-mchat-663265a8351231c428318a8f) by L-AI.
-
 """
 
-
 if not torch.cuda.is_available():
     DESCRIPTION += "\n<p>Running on CPU! This demo does not work on CPU.</p>"
 
-
-if torch.cuda.is_available():
-    model_id = "Artples/L-MChat-7b"
-    model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
-    tokenizer = AutoTokenizer.from_pretrained(model_id)
-    tokenizer.use_default_system_prompt = False
-
+model_options = {
+    "Fast-Model": "Artples/L-MChat-Small",
+    "Quality-Model": "Artples/L-MChat-7b"
+}
 
 @spaces.GPU(enable_queue=True, duration=90)
 def generate(
     message: str,
+    model_choice: str,
     chat_history: list[tuple[str, str]],
     system_prompt: str,
     max_new_tokens: int = 1024,
@@ -41,6 +36,11 @@ def generate(
     top_k: int = 50,
     repetition_penalty: float = 1.2,
 ) -> Iterator[str]:
+    model_id = model_options[model_choice]
+    model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
+    tokenizer = AutoTokenizer.from_pretrained(model_id)
+    tokenizer.use_default_system_prompt = False
+
     conversation = []
     if system_prompt:
         conversation.append({"role": "system", "content": system_prompt})
@@ -48,87 +48,40 @@ def generate(
         conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
     conversation.append({"role": "user", "content": message})
 
-    input_ids = tokenizer.apply_chat_template(conversation, return_tensors="pt", add_generation_prompt=True)
-    if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
-        input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
-        gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
-    input_ids = input_ids.to(model.device)
+    input_ids = tokenizer(conversation, return_tensors="pt", padding=True, truncation=True)
+    if input_ids['input_ids'].shape[1] > MAX_INPUT_TOKEN_LENGTH:
+        input_ids['input_ids'] = input_ids['input_ids'][:, -MAX_INPUT_TOKEN_LENGTH:]
 
-    streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
-    generate_kwargs = dict(
-        {"input_ids": input_ids},
-        streamer=streamer,
-        max_new_tokens=max_new_tokens,
-        do_sample=True,
+    outputs = model.generate(
+        **input_ids,
+        max_length=input_ids['input_ids'].shape[1] + max_new_tokens,
         top_p=top_p,
         top_k=top_k,
         temperature=temperature,
-        num_beams=1,
-        repetition_penalty=repetition_penalty,
+        num_return_sequences=1,
+        repetition_penalty=repetition_penalty
     )
-    t = Thread(target=model.generate, kwargs=generate_kwargs)
-    t.start()
 
-    outputs = []
-    for text in streamer:
-        outputs.append(text)
-        yield "".join(outputs)
+    generated_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    yield generated_text
 
-
-chat_interface = gr.ChatInterface(
-    theme='ehristoforu/RE_Theme',
+chat_interface = gr.Interface(
     fn=generate,
-    additional_inputs=[
-        gr.Textbox(label="System prompt", lines=6),
-        gr.Slider(
-            label="Max new tokens",
-            minimum=1,
-            maximum=MAX_MAX_NEW_TOKENS,
-            step=1,
-            value=DEFAULT_MAX_NEW_TOKENS,
-        ),
-        gr.Slider(
-            label="Temperature",
-            minimum=0.1,
-            maximum=4.0,
-            step=0.1,
-            value=0.6,
-        ),
-        gr.Slider(
-            label="Top-p (nucleus sampling)",
-            minimum=0.05,
-            maximum=1.0,
-            step=0.05,
-            value=0.9,
-        ),
-        gr.Slider(
-            label="Top-k",
-            minimum=1,
-            maximum=1000,
-            step=1,
-            value=50,
-        ),
-        gr.Slider(
-            label="Repetition penalty",
-            minimum=1.0,
-            maximum=2.0,
-            step=0.05,
-            value=1.2,
-        ),
-    ],
-    stop_btn=None,
-    examples=[
-        ["Hello there! How are you doing?"],
-        ["Can you explain briefly to me what is the Python programming language?"],
-        ["Explain the plot of Cinderella in a sentence."],
-        ["How many hours does it take a man to eat a Helicopter?"],
-        ["Write a 100-word article on 'Benefits of Open-Source in AI research'"],
+    inputs=[
+        gr.Textbox(lines=2, placeholder="Type your message here..."),
+        gr.Dropdown(label="Choose Model", choices=list(model_options.keys())),
+        gr.State(label="Chat History", default=[]),
+        gr.Textbox(label="System Prompt", lines=6, placeholder="Enter system prompt if any..."),
+        gr.Slider(label="Max new tokens", minimum=1, maximum=MAX_MAX_NEW_TOKENS, step=1, value=DEFAULT_MAX_NEW_TOKENS),
+        gr.Slider(label="Temperature", minimum=0.1, maximum=4.0, step=0.1, value=0.1),
+        gr.Slider(label="Top-p (nucleus sampling)", minimum=0.05, maximum=1.0, step=0.05, value=0.9),
+        gr.Slider(label="Top-k", minimum=1, maximum=1000, step=1, value=50),
+        gr.Slider(label="Repetition penalty", minimum=1.0, maximum=2.0, step=0.05, value=1.2),
     ],
+    outputs=[gr.Textbox(label="Response")],
+    theme="default",
+    description=DESCRIPTION
 )
 
-with gr.Blocks(css="style.css") as demo:
-    gr.Markdown(DESCRIPTION)
-    chat_interface.render()
-
 if __name__ == "__main__":
-    demo.queue(max_size=20).launch()
+    chat_interface.launch()