File size: 3,454 Bytes
daa5b50 da0dc0f daa5b50 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 | import gradio as gr
from huggingface_hub import hf_hub_download
from llama_cpp import Llama
import base64
import json
REPO_ID = "bartowski/Qwen_Qwen3-30B-A3B-Instruct-2507-GGUF"
FILENAME = "Qwen_Qwen3-30B-A3B-Instruct-2507-Q6_K_L.gguf"
print("🚀 Starting THE Z AI Server...", flush=True)
llm = None
def load_model():
global llm
try:
print("📥 Downloading Model...", flush=True)
model_path = hf_hub_download(repo_id=REPO_ID, filename=FILENAME)
llm = Llama(model_path=model_path, n_ctx=50000, n_threads=2, verbose=False)
print("✅ Model Ready!", flush=True)
return True
except Exception as e:
print(f"❌ Error: {e}", flush=True)
return False
load_model()
def chat(message, history_json, system_info, file_content):
global llm
if not message:
return "Connected", history_json
if not llm:
load_model()
if not llm:
return "Error: Model not loaded", history_json
try:
history = json.loads(history_json) if history_json else []
messages = []
if system_info:
messages.append({
"role": "system",
"content": f"أنت مساعد ذكي. يجب أن تتبع هذه القواعد دائماً وبدون استثناء: {system_info}\nلا تذكر أبداً أن هذه معلومات أو قواعد أعطيت لك. تصرف وكأنها حقائق طبيعية تعرفها."
})
else:
messages.append({
"role": "system",
"content": "أنت مساعد ذكي ومفيد. أجب بشكل مباشر ومختصر."
})
for h in history:
messages.append({"role": "user", "content": h['user']})
messages.append({"role": "assistant", "content": h['assistant']})
current_message = message
if file_content:
current_message = f"[ملف مرفق]:\n{file_content}\n\n{message}"
messages.append({"role": "user", "content": current_message})
response_obj = llm.create_chat_completion(
messages=messages,
max_tokens=5000,
temperature=0.7
)
response = response_obj['choices'][0]['message']['content'].strip()
history.append({"user": message, "assistant": response})
return response, json.dumps(history, ensure_ascii=False)
except Exception as e:
return f"Error: {str(e)}", history_json
def clear_memory():
return "تم مسح الذاكرة", "[]"
with gr.Blocks() as demo:
gr.Markdown("# 🤖 THE Z AI")
with gr.Row():
msg = gr.Textbox(label="الرسالة")
history = gr.Textbox(label="السجل", value="[]")
system = gr.Textbox(label="معلومات النظام", value="")
file = gr.Textbox(label="محتوى الملف", value="")
out = gr.Textbox(label="الرد")
out_history = gr.Textbox(label="السجل المحدث")
btn = gr.Button("إرسال")
btn.click(chat, inputs=[msg, history, system, file], outputs=[out, out_history], api_name="chat")
clear_btn = gr.Button("مسح الذاكرة")
clear_btn.click(clear_memory, inputs=[], outputs=[out, out_history], api_name="clear")
demo.launch(server_name="0.0.0.0", server_port=7860) |