Download app.py from lahyk/myyycode: direct link, hf CLI and curl.
- Browser
- Download file 1.55 kB
-
https://huggingface.co/spaces/lahyk/myyycode/resolve/main/app.py
- Command line
-
hf download hf://spaces/lahyk/myyycode/app.py
-
curl -L -o app.py https://huggingface.co/spaces/lahyk/myyycode/resolve/main/app.py
1.55 kB
| import gradio as gr | |
| import spaces | |
| import torch | |
| from transformers import AutoProcessor, AutoModelForMultimodalLM | |
| MODEL_ID = "google/gemma-4-E4B-it" | |
| processor = AutoProcessor.from_pretrained(MODEL_ID) | |
| model = AutoModelForMultimodalLM.from_pretrained( | |
| MODEL_ID, | |
| dtype="auto", | |
| device_map="auto", | |
| ) | |
| def chat(message, history): | |
| messages = [] | |
| for user, assistant in history: | |
| messages.append({ | |
| "role": "user", | |
| "content": user | |
| }) | |
| messages.append({ | |
| "role": "assistant", | |
| "content": assistant | |
| }) | |
| messages.append({ | |
| "role": "user", | |
| "content": message | |
| }) | |
| inputs = processor.apply_chat_template( | |
| messages, | |
| tokenize=True, | |
| return_dict=True, | |
| return_tensors="pt", | |
| add_generation_prompt=True, | |
| enable_thinking=False, | |
| ).to(model.device) | |
| input_len = inputs["input_ids"].shape[-1] | |
| with torch.inference_mode(): | |
| outputs = model.generate( | |
| **inputs, | |
| max_new_tokens=512, | |
| do_sample=True, | |
| temperature=0.7, | |
| top_p=0.95, | |
| ) | |
| response = processor.decode( | |
| outputs[0][input_len:], | |
| skip_special_tokens=False | |
| ) | |
| response = processor.parse_response(response) | |
| return response | |
| demo = gr.ChatInterface( | |
| fn=chat, | |
| title="Gemma 4 E4B", | |
| textbox=gr.Textbox( | |
| placeholder="Напиши что-нибудь...", | |
| container=True | |
| ), | |
| ) | |
| demo.launch() |