Upgrade gradio to 4.17

fastchat/llm_judge/qa_browser.py

@@ -36,7 +36,7 @@
 
 def display_question(category_selector, request: gr.Request):
     choices = category_selector_map[category_selector]
-    return gr.Dropdown.update(
+    return gr.Dropdown(
         value=choices[0],
         choices=choices,
     )
@@ -413,6 +413,8 @@ def build_demo():
     ) = load_pairwise_model_judgments(pairwise_model_judgment_file)
 
     demo = build_demo()
-    demo.queue(concurrency_count=10, status_update_rate=10, api_open=False).launch(
+    demo.queue(
+        default_concurrency_limit=10, status_update_rate=10, api_open=False
+    ).launch(
         server_name=args.host, server_port=args.port, share=args.share, max_threads=200
     )

fastchat/model/model_adapter.py

@@ -64,10 +64,12 @@
     "gpt-3.5-turbo-0301",
     "gpt-3.5-turbo-0613",
     "gpt-3.5-turbo-1106",
+    "gpt-3.5-turbo-0125",
     "gpt-4",
     "gpt-4-0314",
     "gpt-4-0613",
     "gpt-4-turbo",
+    "gpt-4-1106-preview",
     "gpt-4-0125-preview",
 )
 

fastchat/model/model_registry.py

@@ -41,12 +41,39 @@ def get_model_info(name: str) -> ModelInfo:
 )
 
 register_model_info(
-    ["mixtral-8x7b-instruct-v0.1", "mistral-medium", "mistral-7b-instruct"],
+    [
+        "mixtral-8x7b-instruct-v0.1",
+        "mistral-medium",
+        "mistral-7b-instruct-v0.2",
+        "mistral-7b-instruct",
+    ],
     "Mixtral of experts",
     "https://mistral.ai/news/mixtral-of-experts/",
     "A Mixture-of-Experts model by Mistral AI",
 )
 
+register_model_info(
+    [
+        "qwen1.5-72b-chat",
+        "qwen1.5-14b-chat",
+        "qwen1.5-7b-chat",
+        "qwen1.5-4b-chat",
+        "qwen1.5-1.8b-chat",
+        "qwen1.5-0.5b-chat",
+        "qwen-14b-chat",
+    ],
+    "Qwen 1.5",
+    "https://qwenlm.github.io/blog/qwen1.5/",
+    "a large language model by Alibaba Cloud",
+)
+
+register_model_info(
+    ["qwen-14b-chat"],
+    "Qwen",
+    "https://huggingface.co/Qwen",
+    "a large language model by Alibaba Cloud",
+)
+
 register_model_info(
     ["bard-feb-2024", "bard-jan-24-gemini-pro"],
     "Bard",
@@ -83,14 +110,20 @@ def get_model_info(name: str) -> ModelInfo:
 )
 
 register_model_info(
-    ["gpt-4-turbo", "gpt-4-0125-preview"],
+    ["gpt-4-turbo", "gpt-4-1106-preview", "gpt-4-0125-preview"],
     "GPT-4-Turbo",
     "https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo",
     "GPT-4-Turbo by OpenAI",
 )
 
 register_model_info(
-    ["gpt-3.5-turbo", "gpt-3.5-turbo-0314", "gpt-3.5-turbo-0613", "gpt-3.5-turbo-1106"],
+    [
+        "gpt-3.5-turbo",
+        "gpt-3.5-turbo-0125",
+        "gpt-3.5-turbo-1106",
+        "gpt-3.5-turbo-0314",
+        "gpt-3.5-turbo-0613",
+    ],
     "GPT-3.5",
     "https://platform.openai.com/docs/models/gpt-3-5",
     "GPT-3.5-Turbo by OpenAI",
@@ -124,6 +157,34 @@ def get_model_info(name: str) -> ModelInfo:
     "Claude Instant by Anthropic",
 )
 
+register_model_info(
+    ["nous-hermes-2-mixtral-8x7b-dpo"],
+    "Nous-Hermes-2-Mixtral-8x7B-DPO",
+    "https://huggingface.co/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO",
+    "Nous Hermes finetuned from Mixtral 8x7B",
+)
+
+register_model_info(
+    ["openchat-3.5-0106", "openchat-3.5"],
+    "OpenChat 3.5",
+    "https://github.com/imoneoi/openchat",
+    "an open model fine-tuned on Mistral-7B using C-RLFT",
+)
+
+register_model_info(
+    ["deepseek-llm-67b-chat"],
+    "DeepSeek LLM",
+    "https://huggingface.co/deepseek-ai/deepseek-llm-67b-chat",
+    "An advanced language model by DeepSeek",
+)
+
+register_model_info(
+    ["stripedhyena-nous-7b"],
+    "StripedHyena-Nous",
+    "https://huggingface.co/togethercomputer/StripedHyena-Nous-7B",
+    "A chat model developed by Together Research and Nous Research.",
+)
+
 register_model_info(
     ["llama2-70b-steerlm-chat"],
     "Llama2-70B-SteerLM-Chat",
@@ -194,13 +255,6 @@ def get_model_info(name: str) -> ModelInfo:
     "an open bilingual dialogue language model by Tsinghua University",
 )
 
-register_model_info(
-    ["openchat-3.5"],
-    "OpenChat 3.5",
-    "https://github.com/imoneoi/openchat",
-    "an open model fine-tuned on Mistral-7B using C-RLFT",
-)
-
 register_model_info(
     ["tenyxchat-7b-v1"],
     "TenyxChat-7B",
@@ -236,13 +290,6 @@ def get_model_info(name: str) -> ModelInfo:
     "The TinyLlama project is an open endeavor to pretrain a 1.1B Llama model on 3 trillion tokens.",
 )
 
-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",

fastchat/serve/api_provider.py

@@ -411,7 +411,7 @@ def mistral_api_stream_iter(model_name, messages, temperature, top_p, max_new_to
 
 
 def nvidia_api_stream_iter(model_name, messages, temp, top_p, max_tokens, api_base):
-    assert model_name in ["llama2-70b-steerlm-chat"]
+    assert model_name in ["llama2-70b-steerlm-chat", "yi-34b-chat"]
 
     api_key = os.environ["NVIDIA_API_KEY"]
     headers = {

fastchat/serve/call_monitor.py

@@ -19,8 +19,11 @@ def __init__(self, log_dir: str):
         self.log_dir = log_dir
         self.model_call = {}
         self.user_call = {}
-        self.model_call_limit_global = {"gpt-4-turbo": 300, "gpt-4-0125-preview": 300}
-        self.model_call_day_limit_per_user = {"gpt-4-turbo": 10}
+        self.model_call_limit_global = {
+            "gpt-4-1106-preview": 300,
+            "gpt-4-0125-preview": 300,
+        }
+        self.model_call_day_limit_per_user = {"gpt-4-1106-preview": 10}
 
     async def update_stats(self, num_file=1) -> None:
         while True: