UI and model change

fastchat/constants.py

@@ -11,11 +11,12 @@
 SERVER_ERROR_MSG = (
     "**NETWORK ERROR DUE TO HIGH TRAFFIC. PLEASE REGENERATE OR REFRESH THIS PAGE.**"
 )
-MODERATION_MSG = "YOUR INPUT VIOLATES OUR CONTENT MODERATION GUIDELINES. PLEASE FIX YOUR INPUT AND TRY AGAIN."
+MODERATION_MSG = "$MODERATION$ YOUR INPUT VIOLATES OUR CONTENT MODERATION GUIDELINES."
 CONVERSATION_LIMIT_MSG = "YOU HAVE REACHED THE CONVERSATION LENGTH LIMIT. PLEASE CLEAR HISTORY AND START A NEW CONVERSATION."
 INACTIVE_MSG = "THIS SESSION HAS BEEN INACTIVE FOR TOO LONG. PLEASE REFRESH THIS PAGE."
+SLOW_MODEL_MSG = "⚠️  Both models will show the responses all at once. Please stay patient as it may take over 30 seconds."
 # Maximum input length
-INPUT_CHAR_LEN_LIMIT = int(os.getenv("FASTCHAT_INPUT_CHAR_LEN_LIMIT", 3072))
+INPUT_CHAR_LEN_LIMIT = int(os.getenv("FASTCHAT_INPUT_CHAR_LEN_LIMIT", 12000))
 # Maximum conversation turns
 CONVERSATION_TURN_LIMIT = 50
 # Session expiration time

fastchat/model/model_registry.py

@@ -28,14 +28,26 @@ def get_model_info(name: str) -> ModelInfo:
         )
 
 
-register_model_info(
-    ["gpt-4"], "ChatGPT-4", "https://openai.com/research/gpt-4", "ChatGPT-4 by OpenAI"
-)
 register_model_info(
     ["gpt-3.5-turbo"],
-    "ChatGPT-3.5",
+    "GPT-3.5",
     "https://openai.com/blog/chatgpt",
-    "ChatGPT-3.5 by OpenAI",
+    "GPT-3.5 by OpenAI",
+)
+register_model_info(
+    ["gpt-3.5-turbo-1106"],
+    "GPT-3.5-Turbo-1106",
+    "https://platform.openai.com/docs/models/gpt-3-5",
+    "GPT-3.5-Turbo-1106 by OpenAI",
+)
+register_model_info(
+    ["gpt-4"], "GPT-4", "https://openai.com/research/gpt-4", "ChatGPT-4 by OpenAI"
+)
+register_model_info(
+    ["gpt-4-turbo"],
+    "GPT-4-Turbo",
+    "https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo",
+    "GPT-4-Turbo by OpenAI",
 )
 register_model_info(
     ["claude-2"],
@@ -61,18 +73,6 @@ def get_model_info(name: str) -> ModelInfo:
     "https://cloud.google.com/vertex-ai/docs/release-notes#May_10_2023",
     "PaLM 2 for Chat (chat-bison@001) by Google",
 )
-register_model_info(
-    ["llama-2-70b-chat", "llama-2-34b-chat", "llama-2-13b-chat", "llama-2-7b-chat"],
-    "Llama 2",
-    "https://ai.meta.com/llama/",
-    "open foundation and fine-tuned chat models by Meta",
-)
-register_model_info(
-    ["codellama-34b-instruct", "codellama-13b-instruct", "codellama-7b-instruct"],
-    "Code Llama",
-    "https://ai.meta.com/blog/code-llama-large-language-model-coding/",
-    "open foundation models for code by Meta",
-)
 register_model_info(
     [
         "vicuna-33b",
@@ -86,6 +86,36 @@ def get_model_info(name: str) -> ModelInfo:
     "https://lmsys.org/blog/2023-03-30-vicuna/",
     "a chat assistant fine-tuned on user-shared conversations by LMSYS",
 )
+register_model_info(
+    ["llama-2-70b-chat", "llama-2-34b-chat", "llama-2-13b-chat", "llama-2-7b-chat"],
+    "Llama 2",
+    "https://ai.meta.com/llama/",
+    "open foundation and fine-tuned chat models by Meta",
+)
+register_model_info(
+    ["mistral-7b-instruct"],
+    "Mistral",
+    "https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1",
+    "a large language model by Mistral AI team",
+)
+register_model_info(
+    ["zephyr-7b-beta", "zephyr-7b-alpha"],
+    "Zephyr",
+    "https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha",
+    "a chatbot fine-tuned from Mistral by Hugging Face",
+)
+register_model_info(
+    ["qwen-14b-chat"],
+    "Qwen",
+    "https://huggingface.co/Qwen/Qwen-14B-Chat",
+    "a large language model by Alibaba Cloud",
+)
+register_model_info(
+    ["codellama-34b-instruct", "codellama-13b-instruct", "codellama-7b-instruct"],
+    "Code Llama",
+    "https://ai.meta.com/blog/code-llama-large-language-model-coding/",
+    "open foundation models for code by Meta",
+)
 register_model_info(
     ["wizardlm-70b", "wizardlm-30b", "wizardlm-13b"],
     "WizardLM",
@@ -153,7 +183,7 @@ def get_model_info(name: str) -> ModelInfo:
     "an Open Assistant for everyone by LAION",
 )
 register_model_info(
-    ["openchat_3.5"],
+    ["openchat-3.5"],
     "OpenChat 3.5",
     "https://github.com/imoneoi/openchat",
     "OpenChat 3.5 is a versatile, open-source language model fine-tuned using C-RLFT",
@@ -313,25 +343,11 @@ def get_model_info(name: str) -> ModelInfo:
     "Vigogne-Chat is a French large language model (LLM) optimized for instruction-following and multi-turn dialogues, developed by Bofeng Huang",
 )
 register_model_info(
-    ["mistral-7b-instruct"],
-    "Mistral",
-    "https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1",
-    "a large language model by Mistral AI team",
-)
-register_model_info(
-    ["deluxe-chat-v1"],
+    ["deluxe-chat-v1", "deluxe-chat-v1.1"],
     "DeluxeChat",
     "",
     "Deluxe Chat",
 )
-
-register_model_info(
-    ["zephyr-7b-alpha"],
-    "Zephyr",
-    "https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha",
-    "a chatbot fine-tuned from Mistral by Hugging Face",
-)
-
 register_model_info(
     [
         "Xwin-LM-7B-V0.1",

fastchat/serve/api_provider.py

@@ -24,6 +24,8 @@ def openai_api_stream_iter(
 
     openai.api_base = api_base or "https://api.openai.com/v1"
     openai.api_key = api_key or os.environ["OPENAI_API_KEY"]
+    if model_name == "gpt-4-turbo":
+        model_name = "gpt-4-1106-preview"
 
     # Make requests
     gen_params = {