Update app.py

app.py

@@ -118,7 +118,7 @@ def process_model(model_id, q_method, use_imatrix, imatrix_q_method, private_rep
             print(f"Model directory contents: {os.listdir(local_dir)}")
 
             conversion_script = "convert_hf_to_gguf.py"
-            fp16_conversion = f"python llama.cpp/{conversion_script} {local_dir} --outtype bf16 --outfile {fp16}"
+            fp16_conversion = f"python llama.cpp/{conversion_script} {local_dir} --outtype f32 --outfile {fp16}"
             result = subprocess.run(fp16_conversion, shell=True, capture_output=True)
             print(result)
             if result.returncode != 0:
@@ -274,7 +274,7 @@ with gr.Blocks(css=css) as demo:
     )
 
     q_method = gr.Dropdown(
-        ["Q2_K", "Q3_K_S", "Q3_K_M", "Q3_K_L", "Q4_0", "Q4_K_S", "Q4_K_M", "Q5_0", "Q5_K_S", "Q5_K_M", "Q6_K", "Q8_0", "F16", "BF16"],
+        ["Q2_K", "Q3_K_S", "Q3_K_M", "Q3_K_L", "Q4_0", "Q4_K_S", "Q4_K_M", "Q5_0", "Q5_K_S", "Q5_K_M", "Q6_K", "Q8_0", "F16", "BF16", "F32"],
         label="Quantization Method",
         info="GGML quantization type",
         value="Q4_K_M",