Update app.py

app.py

@@ -1,20 +1,21 @@
+from llama_cpp import Llama
 import os
-from transformers import pipeline
 
-# Force cache location BEFORE importing transformers
-os.environ['TRANSFORMERS_CACHE'] = '/tmp/cache'
-os.environ['HF_HOME'] = '/tmp/cache'
-os.environ['HF_DATASETS_CACHE'] = '/tmp/cache'
-os.environ['HUGGINGFACE_HUB_CACHE'] = '/tmp/cache'
+# Configure cache
+os.environ['GGUF_CACHE'] = '/tmp/gguf_cache'
+os.makedirs('/tmp/gguf_cache', exist_ok=True)
 
-# Now import pipeline
-model = pipeline(
-    "text-generation", 
-    model="distilgpt2",  # Smaller model for reliability
-    device=-1  # Force CPU
+# Load GGUF model
+llm = Llama(
+    model_path="ninja-v1-nsfw-rp.gguf",
+    n_ctx=2048,  # Context window
+    n_threads=4  # CPU threads
 )
 
-def generate_text(prompt, max_length=100):
-    """Generate text from a prompt"""
-    output = model(prompt, max_length=max_length)
-    return output[0]["generated_text"]
+def generate_text(prompt, max_tokens=560):
+    output = llm.create_chat_completion(
+        messages=[{"role": "user", "content": prompt}],
+        max_tokens=max_tokens,
+        temperature=0.7
+    )
+    return output['choices'][0]['message']['content']