Merge pull request #2945 from zRzRzRzRzRzRzR/dev

更新zhipuai请求方式

README.md

@@ -193,7 +193,7 @@ $ python startup.py -a
 [![Telegram](https://img.shields.io/badge/Telegram-2CA5E0?style=for-the-badge&logo=telegram&logoColor=white "langchain-chatglm")](https://t.me/+RjliQ3jnJ1YyN2E9)
 
 ### 项目交流群
-<img src="img/qr_code_88.jpg" alt="二维码" width="300" />
+<img src="img/qr_code_89.jpg" alt="二维码" width="300" />
 
 🎉 Langchain-Chatchat 项目微信交流群，如果你也对本项目感兴趣，欢迎加入群聊参与讨论交流。
 

README_en.md

@@ -192,7 +192,7 @@ please refer to the [Wiki](https://github.com/chatchat-space/Langchain-Chatchat/
 
 ### WeChat Group
 
-<img src="img/qr_code_88.jpg" alt="二维码" width="300" height="300" />
+<img src="img/qr_code_89.jpg" alt="二维码" width="300" height="300" />
 
 ### WeChat Official Account
 

README_ja.md

@@ -185,7 +185,7 @@ $ python startup.py -a
 
 ### WeChat グループ
 
-<img src="img/qr_code_88.jpg" alt="二维码" width="300" height="300" />
+<img src="img/qr_code_89.jpg" alt="二维码" width="300" height="300" />
 
 ### WeChat 公式アカウント
 

configs/model_config.py.example

@@ -150,12 +150,16 @@ MODEL_PATH = {
         "m3e-small": "moka-ai/m3e-small",
         "m3e-base": "moka-ai/m3e-base",
         "m3e-large": "moka-ai/m3e-large",
+
         "bge-small-zh": "BAAI/bge-small-zh",
         "bge-base-zh": "BAAI/bge-base-zh",
         "bge-large-zh": "BAAI/bge-large-zh",
         "bge-large-zh-noinstruct": "BAAI/bge-large-zh-noinstruct",
         "bge-base-zh-v1.5": "BAAI/bge-base-zh-v1.5",
         "bge-large-zh-v1.5": "BAAI/bge-large-zh-v1.5",
+
+        "bge-m3": "BAAI/bge-m3",
+
         "piccolo-base-zh": "sensenova/piccolo-base-zh",
         "piccolo-large-zh": "sensenova/piccolo-large-zh",
         "nlp_gte_sentence-embedding_chinese-large": "damo/nlp_gte_sentence-embedding_chinese-large",
@@ -181,6 +185,14 @@ MODEL_PATH = {
         "Qwen-14B-Chat": "Qwen/Qwen-14B-Chat",
         "Qwen-72B-Chat": "Qwen/Qwen-72B-Chat",
 
+        # Qwen1.5 模型 VLLM可能出现问题
+        "Qwen1.5-0.5B-Chat": "Qwen/Qwen1.5-0.5B-Chat",
+        "Qwen1.5-1.8B-Chat": "Qwen/Qwen1.5-1.8B-Chat",
+        "Qwen1.5-4B-Chat": "Qwen/Qwen1.5-4B-Chat",
+        "Qwen1.5-7B-Chat": "Qwen/Qwen1.5-7B-Chat",
+        "Qwen1.5-14B-Chat": "Qwen/Qwen1.5-14B-Chat",
+        "Qwen1.5-72B-Chat": "Qwen/Qwen1.5-72B-Chat",
+
         "baichuan-7b-chat": "baichuan-inc/Baichuan-7B-Chat",
         "baichuan-13b-chat": "baichuan-inc/Baichuan-13B-Chat",
         "baichuan2-7b-chat": "baichuan-inc/Baichuan2-7B-Chat",

configs/server_config.py.example

@@ -90,13 +90,12 @@ FSCHAT_MODEL_WORKERS = {
         # 'disable_log_requests': False
 
     },
-    "Qwen-1_8B-Chat": {
-        "device": "cpu",
-    },
     "chatglm3-6b": {
         "device": "cuda",
     },
-
+    "Qwen1.5-0.5B-Chat": {
+        "device": "cuda",
+    },
     # 以下配置可以不用修改，在model_config中设置启动的模型
     "zhipu-api": {
         "port": 21001,

requirements.txt

@@ -2,7 +2,7 @@ torch==2.1.2
 torchvision==0.16.2
 torchaudio==2.1.2
 xformers==0.0.23.post1
-transformers==4.37.1
+transformers==4.37.2
 sentence_transformers==2.2.2
 langchain==0.0.354
 langchain-experimental==0.0.47

requirements_api.txt

@@ -2,7 +2,7 @@ torch~=2.1.2
 torchvision~=0.16.2
 torchaudio~=2.1.2
 xformers>=0.0.23.post1
-transformers==4.37.1
+transformers==4.37.2
 sentence_transformers==2.2.2
 langchain==0.0.354
 langchain-experimental==0.0.47

server/knowledge_base/kb_service/milvus_kb_service.py

@@ -50,7 +50,7 @@ class MilvusKBService(KBService):
 
     def _load_milvus(self):
         self.milvus = Milvus(embedding_function=EmbeddingsFunAdapter(self.embed_model),
-                             collection_name=self.kb_name, 
+                             collection_name=self.kb_name,
                              connection_args=kbs_config.get("milvus"),
                              index_params=kbs_config.get("milvus_kwargs")["index_params"],
                              search_params=kbs_config.get("milvus_kwargs")["search_params"]
@@ -89,6 +89,14 @@ class MilvusKBService(KBService):
         if self.milvus.col:
             self.milvus.col.delete(expr=f'pk in {id_list}')
 
+        # Issue 2846, for windows
+        # if self.milvus.col:
+        #     file_path = kb_file.filepath.replace("\\", "\\\\")
+        #     file_name = os.path.basename(file_path)
+        #     id_list = [item.get("pk") for item in
+        #                self.milvus.col.query(expr=f'source == "{file_name}"', output_fields=["pk"])]
+        #     self.milvus.col.delete(expr=f'pk in {id_list}')
+
     def do_clear_vs(self):
         if self.milvus.col:
             self.do_drop_kb()

server/model_workers/xinghuo.py

@@ -13,7 +13,6 @@ async def request(appid, api_key, api_secret, Spark_url, domain, question, tempe
     wsParam = SparkApi.Ws_Param(appid, api_key, api_secret, Spark_url)
     wsUrl = wsParam.create_url()
     data = SparkApi.gen_params(appid, domain, question, temperature, max_token)
-    print(data)
     async with websockets.connect(wsUrl) as ws:
         await ws.send(json.dumps(data, ensure_ascii=False))
         finish = False

server/model_workers/zhipu.py

@@ -1,13 +1,21 @@
+from contextlib import contextmanager
+
+import httpx
 from fastchat.conversation import Conversation
+from httpx_sse import EventSource
+
 from server.model_workers.base import *
 from fastchat import conversation as conv
 import sys
-from typing import List, Dict, Iterator, Literal
-from configs import logger, log_verbose
-import requests
+from typing import List, Dict, Iterator, Literal, Any
 import jwt
 import time
-import json
+
+
+@contextmanager
+def connect_sse(client: httpx.Client, method: str, url: str, **kwargs: Any):
+    with client.stream(method, url, **kwargs) as response:
+        yield EventSource(response)
 
 
 def generate_token(apikey: str, exp_seconds: int):
@@ -37,7 +45,7 @@ class ChatGLMWorker(ApiModelWorker):
             model_names: List[str] = ["zhipu-api"],
             controller_addr: str = None,
             worker_addr: str = None,
-            version: Literal["chatglm_turbo"] = "chatglm_turbo",
+            version: Literal["glm-4"] = "glm-4",
             **kwargs,
     ):
         kwargs.update(model_names=model_names, controller_addr=controller_addr, worker_addr=worker_addr)
@@ -59,25 +67,25 @@ class ChatGLMWorker(ApiModelWorker):
             "temperature": params.temperature,
             "stream": False
         }
+
         url = "https://open.bigmodel.cn/api/paas/v4/chat/completions"
-        response = requests.post(url, headers=headers, json=data)
-        # for chunk in response.iter_lines():
-        #     if chunk:
-        #         chunk_str = chunk.decode('utf-8')
-        #         json_start_pos = chunk_str.find('{"id"')
-        #         if json_start_pos != -1:
-        #             json_str = chunk_str[json_start_pos:]
-        #             json_data = json.loads(json_str)
-        #             for choice in json_data.get('choices', []):
-        #                 delta = choice.get('delta', {})
-        #                 content = delta.get('content', '')
-        #                 yield {"error_code": 0, "text": content}
-        ans = response.json()
-        content = ans["choices"][0]["message"]["content"]
-        yield {"error_code": 0, "text": content}
+        with httpx.Client(headers=headers) as client:
+            response = client.post(url, json=data)
+            response.raise_for_status()
+            chunk = response.json()
+            print(chunk)
+            yield {"error_code": 0, "text": chunk["choices"][0]["message"]["content"]}
+
+            # with connect_sse(client, "POST", url, json=data) as event_source:
+            #     for sse in event_source.iter_sse():
+            #         chunk = json.loads(sse.data)
+            #         if len(chunk["choices"]) != 0:
+            #             text += chunk["choices"][0]["delta"]["content"]
+            #             yield {"error_code": 0, "text": text}
+
+
 
     def get_embeddings(self, params):
-        # 临时解决方案，不支持embedding
         print("embedding")
         print(params)