both models working

app.py

@@ -11,7 +11,8 @@ from share_btn import community_icon_html, loading_icon_html, share_js, share_bt
 
 HF_TOKEN = os.environ.get("HF_TOKEN", None)
 
-API_URL = "https://api-inference.huggingface.co/models/Phind/Phind-CodeLlama-34B-v2"
+API_URL = "https://api-inference.huggingface.co/models/"
+model_id_1, model_id_2 = "Phind/Phind-CodeLlama-34B-v2", "WizardLM/WizardCoder-Python-34B-V1.0"
 
 FIM_PREFIX = "<PRE> "
 FIM_MIDDLE = " <MID>"
@@ -35,15 +36,13 @@ theme = gr.themes.Monochrome(
     ],
 )
 
-client = Client(
-    API_URL,
-    headers={"Authorization": f"Bearer {HF_TOKEN}"},
-)
-
-
 def generate(
-    prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
+    model_id, prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0,
 ):
+    client = Client(
+        f"{API_URL}{model_id}",
+        headers={"Authorization": f"Bearer {HF_TOKEN}"},
+    )
 
     temperature = float(temperature)
     if temperature < 1e-2:
@@ -94,7 +93,7 @@ def generate(
     return output
 
 def generate_both(prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0):
-    generator_1, generator_2 = generate(prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0), generate(prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0)
+    generator_1, generator_2 = generate(model_id_1, prompt, temperature, max_new_tokens, top_p, repetition_penalty), generate(model_id_2, prompt, temperature, max_new_tokens, top_p, repetition_penalty)
     output_1, output_2 = "", ""
     output_1_end, output_2_end = False, False
 
@@ -163,8 +162,8 @@ with gr.Blocks(theme=theme, analytics_enabled=False, css=css) as demo:
                     elem_id="q-input",
                 )
                 submit = gr.Button("Generate", variant="primary")
-                output_1 = gr.Code(elem_id="q-output", lines=30, label="Output")
-                output_2 = gr.Code(elem_id="q-output", lines=30, label="Output")
+                output_1 = gr.Code(elem_id="q-output", lines=30, label=f"{model_id_1} Output")
+                output_2 = gr.Code(elem_id="q-output", lines=30, label=f"{model_id_2} Output")
                 with gr.Row():
                     with gr.Column():
                         with gr.Accordion("Advanced settings", open=False):