Update app.py

app.py

@@ -1,62 +1,27 @@
 import gradio as gr
-from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
-import torch
+from huggingface_hub import InferenceClient
+import os
 
-model_name = "DavidAU/Qwen3.5-4B-Claude-4.6-OS-Auto-Variable-HERETIC-UNCENSORED-THINKING"
+# Token dibaca dari secret, bukan hardcode
+token = os.environ.get("HF_TOKEN")
 
-bnb_config = BitsAndBytesConfig(
-    load_in_4bit=True,
-    bnb_4bit_compute_dtype=torch.float16,
-    bnb_4bit_use_double_quant=True,
-    bnb_4bit_quant_type="nf4"
+client = InferenceClient(
+    model="NousResearch/Hermes-3-Llama-3.1-8B",
+    token=token
 )
 
-tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained(
-    model_name,
-    quantization_config=bnb_config,
-    device_map="auto",
-    low_cpu_mem_usage=True,
-    trust_remote_code=True
-)
-model.eval()
-
-def generate_code(prompt):
-    formatted = f"### Instruction:\n{prompt}\n### Response:\n"
-    
-    inputs = tokenizer(
-        formatted,
-        return_tensors="pt",
-        truncation=True,
-        max_length=1024
+def generate(prompt):
+    response = client.chat_completion(
+        messages=[{"role": "user", "content": prompt}],
+        max_tokens=512,
+        temperature=0.7
     )
-    
-    input_ids = inputs["input_ids"].to(model.device)
-    attention_mask = inputs["attention_mask"].to(model.device)
-    
-    with torch.no_grad():
-        output = model.generate(
-            input_ids=input_ids,
-            attention_mask=attention_mask,
-            max_new_tokens=512,
-            do_sample=True,
-            temperature=0.7,
-            pad_token_id=tokenizer.eos_token_id,
-            eos_token_id=tokenizer.eos_token_id
-        )
-    
-    new_tokens = output[0][input_ids.shape[-1]:]
-    result = tokenizer.decode(new_tokens, skip_special_tokens=True)
-    return result.strip()
+    return response.choices[0].message.content
 
 gr.Interface(
-    fn=generate_code,
-    inputs=gr.Textbox(
-        lines=4,
-        label="Request kode kamu",
-        placeholder="Contoh: buatkan fungsi Python untuk REST API"
-    ),
-    outputs=gr.Code(label="Hasil kode", language="python"),
-    title="Coding Assistant - Uncensored",
-    description="Powered by Qwen3.5-4B Uncensored"
+    fn=generate,
+    inputs=gr.Textbox(lines=4, label="Tanya apa saja"),
+    outputs=gr.Textbox(label="Jawaban"),
+    title="AI Assistant - Uncensored",
+    description="Powered by HF Inference API"
 ).launch()