增加加了博物馆展品清单数据库及对前端获取展品清单、展品详细的接口，增加了QWenOmni多模态大模型的支 持（主要为了测试）,增加了本地部署大模型支持（主要为了测试，在autoDL上），修正了TTS生成和返回前端的逻辑与参数，增加了判断用户问题有没有在知识库中检索到相关片段、如果没有则直接返回并提示未包含

rag/llm/chat_model.py

@@ -127,6 +127,98 @@ class DeepSeekChat(Base):
         super().__init__(key, model_name, base_url)
 
 
+class LocalLLMChat(Base):
+    def __init__(self, key, model_name="Qwen2.5-7B", base_url="http://106.52.71.204:9483/v1"):
+        if not base_url: base_url = "http://106.52.71.204:9483/v1"
+        super().__init__(key, model_name, base_url)
+
+class QWenOmniChat(Base):
+    def __init__(self, key, model_name="qwen-omni-turbo", base_url="https://dashscope.aliyuncs.com/compatible-mode/v1"):
+        if not base_url: base_url = "https://dashscope.aliyuncs.com/compatible-mode/v1"
+        super().__init__(key, model_name, base_url)
+    def chat(self, system, history, gen_conf):
+        if system:
+            history.insert(0, {"role": "system", "content": system})
+        ans = ""
+        total_tokens = 0
+        try:
+            response = self.client.chat.completions.create(
+                model = self.model_name,
+                messages=history,
+                stream=True,
+                # 设置输出数据的模态，当前支持两种：["text","audio"]、["text"]
+                # modalities=["text", "audio"],
+                # audio={"voice": "Cherry", "format": "wav"},
+                # stream 必须设置为 True，否则会报错
+                # stream_options={"include_usage": True},
+                **gen_conf
+            )
+            for resp in response:
+                if not resp.choices: continue
+                if not resp.choices[0].delta.content:
+                    resp.choices[0].delta.content = ""
+                ans += resp.choices[0].delta.content
+
+                if not hasattr(resp, "usage") or not resp.usage:
+                    total_tokens = (
+                                total_tokens
+                                + num_tokens_from_string(resp.choices[0].delta.content)
+                        )
+                elif isinstance(resp.usage, dict):
+                    total_tokens = resp.usage.get("total_tokens", total_tokens)
+                else: total_tokens = resp.usage.total_tokens
+
+                if resp.choices[0].finish_reason == "length":
+                    ans += "...\nFor the content length reason, it stopped, continue?" if is_english(
+                        [ans]) else "······\n由于长度的原因，回答被截断了，要继续吗？"
+                    break  # 如果达到长度限制，可以跳出循环
+        except openai.APIError as e:
+            ans= ans + "\n**ERROR**: " + str(e)
+
+        return ans, total_tokens
+
+
+    def chat_streamly(self, system, history, gen_conf):
+        # logging.info(f"chat_streamly :{gen_conf}")
+        if system :
+            history.insert(0, {"role": "system", "content":
+                               [{"type":"text","text": system}]})
+        ans = ""
+        total_tokens = 0
+        try:
+            response = self.client.chat.completions.create(
+                model=self.model_name,
+                messages=history,
+                stream=True,
+                #**gen_conf
+            )
+            for resp in response:
+                if not resp.choices: continue
+                if not resp.choices[0].delta.content:
+                    resp.choices[0].delta.content = ""
+                ans += resp.choices[0].delta.content
+
+                if not hasattr(resp, "usage") or not resp.usage:
+                    total_tokens = (
+                                total_tokens
+                                + num_tokens_from_string(resp.choices[0].delta.content)
+                        )
+                elif isinstance(resp.usage, dict):
+                    total_tokens = resp.usage.get("total_tokens", total_tokens)
+                else: total_tokens = resp.usage.total_tokens
+
+                if resp.choices[0].finish_reason == "length":
+                    ans += "...\nFor the content length reason, it stopped, continue?" if is_english(
+                        [ans]) else "······\n由于长度的原因，回答被截断了，要继续吗？"
+                yield ans
+
+        except openai.APIError as e:
+            yield ans + "\n**ERROR**: " + str(e)
+
+        yield total_tokens
+
+
+
 class AzureChat(Base):
     def __init__(self, key, model_name, **kwargs):
         api_key = json.loads(key).get('api_key', '')