Implement Cohere API

app.py

@@ -1,11 +1,15 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
+import os
+import cohere
 
 """
 For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
 """
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
+COHERE_API_KEY = os.getenv("COHERE_API_KEY")
+client_cohere = cohere.Client(COHERE_API_KEY)
+COHERE_MODEL = "command-r-plus"
 
 def respond(
     message,
@@ -14,6 +18,7 @@ def respond(
     max_tokens,
     temperature,
     top_p,
+    use_cohere_api,
 ):
     messages = [{"role": "system", "content": system_message}]
 
@@ -27,17 +32,26 @@ def respond(
 
     response = ""
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
+    if not use_cohere_api:
+        cohere_response = client_cohere.chat(
+            message=message,
+            model=COHERE_MODEL,
+            temperature=temperature,
+            max_tokens=max_tokens
+        )
+        response = cohere_response.text
         yield response
+    else:
+        for message in client.chat_completion(
+            messages,
+            max_tokens=max_tokens,
+            temperature=temperature,
+            top_p=top_p,
+        ):
+            token = message.choices[0].delta.content
+
+            response += token
+            yield response
 
 
 """
@@ -56,6 +70,7 @@ demo = gr.ChatInterface(
             step=0.05,
             label="Top-p (nucleus sampling)",
         ),
+        gr.Checkbox(label="Use Cohere API", default=False),
     ],
 )
 

requirements.txt

@@ -1 +1,2 @@
-huggingface_hub==0.25.2
+huggingface_hub==0.25.2
+cohere