fix

docs/source/Instruction/命令行参数.md

@@ -361,6 +361,7 @@ RLHF参数继承于[训练参数](#训练参数)
 App参数继承于[部署参数](#部署参数), [Web-UI参数](#Web-UI参数)
 - base_url: 模型部署的base_url，例如`http://localhost:8000/v1`。默认为`None`
 - studio_title: studio的标题。默认为None，设置为模型名
+- is_multimodal: 是否启动多模态版本的app。默认为None，自动根据model判断，若无法判断，设置为False
 - lang: 覆盖Web-UI参数，默认为'en'
 
 ### 评测参数

docs/source_en/Instruction/Command-line-parameters.md

@@ -361,6 +361,7 @@ App parameters inherit from [deployment arguments](#deployment-arguments) and [W
 
 - base_url: Base URL for the model deployment, for example, `http://localhost:8000/v1`. Default is None.
 - studio_title: Title of the studio. Default is None, set to the model name.
+- is_multimodal: Whether to launch the multimodal version of the app. Defaults to None, automatically determined based on the model; if it cannot be determined, set to False.
 - lang: Overrides the Web-UI Arguments, default is 'en'.
 
 ### Evaluation Arguments

swift/llm/app/app.py

@@ -22,8 +22,9 @@ def run(self):
             base_url = base_url or args.base_url
             demo = build_ui(
                 base_url,
+                args.model_suffix,
                 stream=args.stream,
-                is_multimodal=args.model_meta.is_multimodal,
+                is_multimodal=args.is_multimodal,
                 studio_title=args.studio_title,
                 lang=args.lang,
                 default_system=args.system)

swift/llm/app/build_ui.py

@@ -18,7 +18,7 @@ def modify_system_session(system: str):
     return system, '', []
 
 
-def _history_to_messages(history: History, system: str):
+def _history_to_messages(history: History, system: Optional[str]):
     messages = []
     if system is not None:
         messages.append({'role': 'system', 'content': system})
@@ -43,12 +43,14 @@ def _history_to_messages(history: History, system: str):
     return messages
 
 
-def model_chat(history: History, system: str, *, client, stream: bool):
+def model_chat(history: History, system: Optional[str], *, client, model: str, stream: bool):
     if history:
         from swift.llm import InferRequest, RequestConfig
 
         messages = _history_to_messages(history, system)
-        gen_or_res = client.infer([InferRequest(messages=messages)], request_config=RequestConfig(stream=stream))
+        gen_or_res = client.infer([InferRequest(messages=messages)],
+                                  request_config=RequestConfig(stream=stream),
+                                  model=model)
         if stream:
             response = ''
             for resp_list in gen_or_res:
@@ -81,6 +83,7 @@ def add_file(history: History, file):
 
 
 def build_ui(base_url: str,
+             model: Optional[str] = None,
              *,
              stream: bool = True,
              is_multimodal: bool = True,
@@ -89,8 +92,8 @@ def build_ui(base_url: str,
              default_system: Optional[str] = None):
     from swift.llm import InferClient
     client = InferClient(base_url=base_url)
-    if studio_title is None:
-        studio_title = client.models[0]
+    model = model or client.models[0]
+    studio_title = studio_title or model
     with gr.Blocks() as demo:
         gr.Markdown(f'<center><font size=8>{studio_title}</center>')
         with gr.Row():
@@ -108,7 +111,7 @@ def build_ui(base_url: str,
             clear_history = gr.Button(locale_mapping['clear_history'][lang])
 
         system_state = gr.State(value=default_system)
-        model_chat_ = partial(model_chat, client=client, stream=stream)
+        model_chat_ = partial(model_chat, client=client, model=model, stream=stream)
 
         upload.upload(add_file, [chatbot, upload], [chatbot])
         textbox.submit(add_text, [chatbot, textbox], [chatbot, textbox]).then(model_chat_, [chatbot, system_state],

swift/llm/argument/app_args.py

@@ -3,6 +3,7 @@
 from typing import Literal, Optional
 
 from swift.utils import find_free_port, get_logger
+from ..model import get_matched_model_meta
 from ..template import get_template_meta
 from .deploy_args import DeployArguments
 from .webui_args import WebUIArguments
@@ -14,18 +15,23 @@
 class AppArguments(WebUIArguments, DeployArguments):
     base_url: Optional[str] = None
     studio_title: Optional[str] = None
+    is_multimodal: Optional[str] = None
 
     lang: Literal['en', 'zh'] = 'en'
 
     def _init_torch_dtype(self) -> None:
         if self.base_url:
+            self.model_meta = get_matched_model_meta(self.model)
             return
         super()._init_torch_dtype()
 
     def __post_init__(self):
         super().__post_init__()
         self.server_port = find_free_port(self.server_port)
-        if self.studio_title is None:
-            self.studio_title = self.model_suffix
-        if self.system is None:
-            self.system = get_template_meta(self.model_meta.template).default_system
+        if self.model_meta:
+            if self.system is None:
+                self.system = get_template_meta(self.model_meta.template).default_system
+            if self.is_multimodal is None:
+                self.is_multimodal = self.model_meta.is_multimodal
+        if self.is_multimodal is None:
+            self.is_multimodal = False

swift/llm/model/__init__.py

@@ -3,7 +3,7 @@
 from .constant import LLMModelType, MLLMModelType, ModelType
 from .model_arch import MODEL_ARCH_MAPPING, ModelArch, ModelKeys, MultiModelKeys, get_model_arch, register_model_arch
 from .register import (MODEL_MAPPING, Model, ModelGroup, ModelMeta, fix_do_sample_warning, get_default_device_map,
-                       get_default_torch_dtype, get_model_info_meta, get_model_name, get_model_tokenizer,
-                       get_model_tokenizer_multimodal, get_model_tokenizer_with_flash_attn, get_model_with_value_head,
-                       load_by_unsloth, register_model)
+                       get_default_torch_dtype, get_matched_model_meta, get_model_info_meta, get_model_name,
+                       get_model_tokenizer, get_model_tokenizer_multimodal, get_model_tokenizer_with_flash_attn,
+                       get_model_with_value_head, load_by_unsloth, register_model)
 from .utils import HfConfigFactory, ModelInfo, git_clone_github, safe_snapshot_download

tests/app/test_app.py

@@ -14,8 +14,11 @@ def test_mllm():
 
 
 def test_audio():
-    from swift.llm import app_main, AppArguments
-    app_main(AppArguments(model='Qwen/Qwen2-Audio-7B-Instruct', stream=True))
+    from swift.llm import AppArguments, app_main, DeployArguments, run_deploy
+    deploy_args = DeployArguments(model='Qwen/Qwen2-Audio-7B-Instruct', infer_backend='pt', verbose=False)
+
+    with run_deploy(deploy_args, return_url=True) as url:
+        app_main(AppArguments(model='Qwen2-Audio-7B-Instruct', base_url=url, stream=True))
 
 
 if __name__ == '__main__':