add falcon-180B-chat

gradio_app/app.py

@@ -165,7 +165,8 @@ with gr.Blocks() as demo:
                 choices=[
                     "gpt-3.5-turbo",
                     "mistralai/Mistral-7B-Instruct-v0.1",
-                    "GeneZC/MiniChat-3B",
+                    "tiiuae/falcon-180B-chat",
+                    # "GeneZC/MiniChat-3B",
                 ],
                 value="gpt-3.5-turbo",
                 label='LLM'

gradio_app/backend/HuggingfaceGenerator.py

@@ -27,7 +27,7 @@ class HuggingfaceGenerator:
             top_p: float = None, repetition_penalty: float = None,
             stream: bool = True,
     ):
-        self.tokenizer = AutoTokenizer.from_pretrained(model_name)
+        self.tokenizer = AutoTokenizer.from_pretrained(model_name, token=HF_TOKEN)
         self.hf_client = InferenceClient(model_name, token=HF_TOKEN)
         self.stream = stream
 

gradio_app/backend/query_llm.py

@@ -53,7 +53,7 @@ def construct_openai_messages(context, history):
 def get_message_constructor(llm_name):
     if llm_name == 'gpt-3.5-turbo':
         return construct_openai_messages
-    if llm_name in ['mistralai/Mistral-7B-Instruct-v0.1', "GeneZC/MiniChat-3B"]:
+    if llm_name in ['mistralai/Mistral-7B-Instruct-v0.1', "tiiuae/falcon-180B-chat", "GeneZC/MiniChat-3B"]:
         return construct_mistral_messages
     raise ValueError('Unknown LLM name')
 
@@ -64,7 +64,7 @@ def get_llm_generator(llm_name):
             model_name=llm_name, max_tokens=512, temperature=0, stream=True
         )
         return cgi.chat_completion
-    if llm_name == 'mistralai/Mistral-7B-Instruct-v0.1':
+    if llm_name == 'mistralai/Mistral-7B-Instruct-v0.1' or llm_name == "tiiuae/falcon-180B-chat":
         hfg = HuggingfaceGenerator(
             model_name=llm_name, temperature=0, max_new_tokens=512,
         )

settings.py

@@ -27,6 +27,7 @@ thresh_distances = {
 
 context_lengths = {
     "mistralai/Mistral-7B-Instruct-v0.1": 4096,
+    "tiiuae/falcon-180B-chat": 2048,
     "GeneZC/MiniChat-3B": 4096,
     "gpt-3.5-turbo": 4096,
     "sentence-transformers/all-MiniLM-L6-v2": 128,