use smaller model (qwen2.5-coder-3b-instruct-q6_k.gguf)
Browse files
    	
        app.py
    CHANGED
    
    | @@ -4,8 +4,8 @@ from llm import load_llm, response_generator | |
| 4 | 
             
            from sql import csv_to_sqlite, run_sql_query
         | 
| 5 |  | 
| 6 |  | 
| 7 | 
            -
            repo_id = " | 
| 8 | 
            -
            filename = " | 
| 9 | 
             
            # repo_id = "Qwen/Qwen2.5-0.5B-Instruct-GGUF"
         | 
| 10 | 
             
            # filename = "qwen2.5-0.5b-instruct-q8_0.gguf"
         | 
| 11 |  | 
|  | |
| 4 | 
             
            from sql import csv_to_sqlite, run_sql_query
         | 
| 5 |  | 
| 6 |  | 
| 7 | 
            +
            repo_id = "Qwen/Qwen2.5-Coder-3B-Instruct-GGUF"
         | 
| 8 | 
            +
            filename = "qwen2.5-coder-3b-instruct-q6_k.gguf"
         | 
| 9 | 
             
            # repo_id = "Qwen/Qwen2.5-0.5B-Instruct-GGUF"
         | 
| 10 | 
             
            # filename = "qwen2.5-0.5b-instruct-q8_0.gguf"
         | 
| 11 |  |