Update version to 0.2.36

fastchat/__init__.py

@@ -1 +1 @@
-__version__ = "0.2.35"
+__version__ = "0.2.36"

fastchat/model/model_registry.py

@@ -291,7 +291,12 @@ def get_model_info(name: str) -> ModelInfo:
 )
 
 register_model_info(
-    ["codellama-34b-instruct", "codellama-13b-instruct", "codellama-7b-instruct"],
+    [
+        "codellama-70b-instruct",
+        "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",
@@ -640,10 +645,10 @@ def get_model_info(name: str) -> ModelInfo:
 
 register_model_info(
     [
-        "llava-v1.6-vicuna-7b",
         "llava-v1.6-vicuna-13b",
-        "llava-v1.5-7b",
+        "llava-v1.6-vicuna-7b",
         "llava-v1.5-13b",
+        "llava-v1.5-7b",
     ],
     "LLaVA",
     "https://github.com/haotian-liu/LLaVA",

fastchat/serve/gradio_block_arena_vision.py

@@ -66,7 +66,7 @@ def build_single_vision_language_model_ui(models, add_promotion_links=False):
         with gr.Column(scale=3):
             textbox = gr.Textbox(
                 show_label=False,
-                placeholder="Enter your prompt here and press ENTER",
+                placeholder="👉 Enter your prompt and press ENTER",
                 container=False,
                 render=False,
                 elem_id="input_box",

fastchat/serve/register_worker.py

@@ -14,13 +14,15 @@
     parser.add_argument("--controller-address", type=str)
     parser.add_argument("--worker-name", type=str)
     parser.add_argument("--check-heart-beat", action="store_true")
+    parser.add_argument("--multimodal", action="store_true")
     args = parser.parse_args()
 
     url = args.controller_address + "/register_worker"
     data = {
         "worker_name": args.worker_name,
         "check_heart_beat": args.check_heart_beat,
         "worker_status": None,
+        "multimodal": args.multimodal,
     }
     r = requests.post(url, json=data)
     assert r.status_code == 200

fastchat/serve/sglang_worker.py

@@ -1,5 +1,5 @@
 """
-A model worker that executes the model based on SGLANG.
+A model worker that executes the model based on SGLang.
 
 Usage:
 python3 -m fastchat.serve.sglang_worker --model-path liuhaotian/llava-v1.5-7b --tokenizer-path llava-hf/llava-1.5-7b-hf --port 30000 --worker-address http://localhost:30000
@@ -121,6 +121,7 @@ async def generate_stream(self, params):
         for i in range(len(split_prompt)):
             prompt.append(split_prompt[i])
             if i < len(images):
+                prompt[-1] = prompt[-1].strip()
                 prompt.append(load_image(images[i]))
 
         state = pipeline.run(

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "fschat"
-version = "0.2.35"
+version = "0.2.36"
 description = "An open platform for training, serving, and evaluating large language model based chatbots."
 readme = "README.md"
 requires-python = ">=3.8"