更新模型配置文件和支持列表

configs/model_config.py.example

@@ -6,9 +6,9 @@ import os
 MODEL_ROOT_PATH = ""
 
 # 选用的 Embedding 名称
-EMBEDDING_MODEL = "bge-large-zh"
+EMBEDDING_MODEL = "bge-large-zh-v1.5"
 
-# Embedding 模型运行设备。设为"auto"会自动检测，也可手动设定为"cuda","mps","cpu"其中之一。
+# Embedding 模型运行设备。设为"auto"会自动检测(会有警告)，也可手动设定为 "cuda","mps","cpu","xpu" 其中之一。
 EMBEDDING_DEVICE = "auto"
 
 # 选用的reranker模型
@@ -26,50 +26,32 @@ EMBEDDING_MODEL_OUTPUT_PATH = "output"
 # 在这里，我们使用目前主流的两个离线模型，其中，chatglm3-6b 为默认加载模型。
 # 如果你的显存不足，可使用 Qwen-1_8B-Chat, 该模型 FP16 仅需 3.8G显存。
 
-# chatglm3-6b输出角色标签<|user|>及自问自答的问题详见项目wiki->常见问题->Q20.
+LLM_MODELS = ["zhipu-api"]
-
-LLM_MODELS = ["chatglm3-6b", "zhipu-api", "openai-api"]  # "Qwen-1_8B-Chat",
-
-# AgentLM模型的名称 (可以不指定，指定之后就锁定进入Agent之后的Chain的模型，不指定就是LLM_MODELS[0])
 Agent_MODEL = None
 
-# LLM 运行设备。设为"auto"会自动检测，也可手动设定为"cuda","mps","cpu"其中之一。
+LLM_DEVICE = "cuda"
-LLM_DEVICE = "auto"
 
-# 历史对话轮数
 HISTORY_LEN = 3
 
-# 大模型最长支持的长度，如果不填写，则使用模型默认的最大长度，如果填写，则为用户设定的最大长度
+MAX_TOKENS = 2048
-MAX_TOKENS = None
 
-# LLM通用对话参数
 TEMPERATURE = 0.7
-# TOP_P = 0.95 # ChatOpenAI暂不支持该参数
 
 ONLINE_LLM_MODEL = {
-    # 线上模型。请在server_config中为每个在线API设置不同的端口
-
     "openai-api": {
-        "model_name": "gpt-3.5-turbo",
+        "model_name": "gpt-4",
         "api_base_url": "https://api.openai.com/v1",
         "api_key": "",
         "openai_proxy": "",
     },
 
-    # 获取api_key请前往https://makersuite.google.com/或者google cloud，使用前先确认网络正常，使用代理请在项目启动（python startup.py -a)环境内设置https_proxy环境变量
+    # 智谱AI API（不支持GLM4，本版本无法兼容，敬请期待0.3.x）具体注册及api key获取请前往 http://open.bigmodel.cn
-    "gemini-api": {
-        "api_key": "", 
-        "provider": "GeminiWorker",
-    },
-
-    # 具体注册及api key获取请前往 http://open.bigmodel.cn
     "zhipu-api": {
         "api_key": "",
-        "version": "chatglm_turbo",  # 可选包括 "chatglm_turbo"
+        "version": "chatglm_turbo",
         "provider": "ChatGLMWorker",
     },
 
-
     # 具体注册及api key获取请前往 https://api.minimax.chat/
     "minimax-api": {
         "group_id": "",
@@ -78,7 +60,6 @@ ONLINE_LLM_MODEL = {
         "provider": "MiniMaxWorker",
     },
 
-
     # 具体注册及api key获取请前往 https://xinghuo.xfyun.cn/
     "xinghuo-api": {
         "APPID": "",
@@ -99,8 +80,8 @@ ONLINE_LLM_MODEL = {
 
     # 火山方舟 API，文档参考 https://www.volcengine.com/docs/82379
     "fangzhou-api": {
-        "version": "chatglm-6b-model",  # 当前支持 "chatglm-6b-model"， 更多的见文档模型支持列表中方舟部分。
+        "version": "chatglm-6b-model",
-        "version_url": "",  # 可以不填写version，直接填写在方舟申请模型发布的API地址
+        "version_url": "",
         "api_key": "",
         "secret_key": "",
         "provider": "FangZhouWorker",
@@ -108,15 +89,15 @@ ONLINE_LLM_MODEL = {
 
     # 阿里云通义千问 API，文档参考 https://help.aliyun.com/zh/dashscope/developer-reference/api-details
     "qwen-api": {
-        "version": "qwen-turbo",  # 可选包括 "qwen-turbo", "qwen-plus"
+        "version": "qwen-max",
-        "api_key": "",  # 请在阿里云控制台模型服务灵积API-KEY管理页面创建
+        "api_key": "",
         "provider": "QwenWorker",
-        "embed_model": "text-embedding-v1" # embedding 模型名称
+        "embed_model": "text-embedding-v1"  # embedding 模型名称
     },
 
     # 百川 API，申请方式请参考 https://www.baichuan-ai.com/home#api-enter
     "baichuan-api": {
-        "version": "Baichuan2-53B",  # 当前支持 "Baichuan2-53B"， 见官方文档。
+        "version": "Baichuan2-53B",
         "api_key": "",
         "secret_key": "",
         "provider": "BaiChuanWorker",
@@ -138,6 +119,11 @@ ONLINE_LLM_MODEL = {
         "secret_key": "",
         "provider": "TianGongWorker",
     },
+    # Gemini API (开发组未测试，由社群提供，只支持pro）
+    "gemini-api": {
+        "api_key": "",
+        "provider": "GeminiWorker",
+    }
 
 }
 
@@ -149,6 +135,7 @@ ONLINE_LLM_MODEL = {
 #           - GanymedeNil/text2vec-large-chinese
 #           - text2vec-large-chinese
 #       2.2 如果以上本地路径不存在，则使用huggingface模型
+
 MODEL_PATH = {
     "embed_model": {
         "ernie-tiny": "nghuyong/ernie-3.0-nano-zh",
@@ -167,7 +154,7 @@ MODEL_PATH = {
         "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-large-zh-v1.5": "/share/home/zyx/Models/bge-large-zh-v1.5",
         "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",
@@ -175,55 +162,55 @@ MODEL_PATH = {
     },
 
     "llm_model": {
-        # 以下部分模型并未完全测试，仅根据fastchat和vllm模型的模型列表推定支持
         "chatglm2-6b": "THUDM/chatglm2-6b",
         "chatglm2-6b-32k": "THUDM/chatglm2-6b-32k",
-
         "chatglm3-6b": "THUDM/chatglm3-6b",
         "chatglm3-6b-32k": "THUDM/chatglm3-6b-32k",
-        "chatglm3-6b-base": "THUDM/chatglm3-6b-base",
 
-        "Qwen-1_8B": "Qwen/Qwen-1_8B",
+        "Llama-2-7b-chat-hf": "meta-llama/Llama-2-7b-chat-hf",
-        "Qwen-1_8B-Chat": "Qwen/Qwen-1_8B-Chat",
+        "Llama-2-13b-chat-hf": "meta-llama/Llama-2-13b-chat-hf",
-        "Qwen-1_8B-Chat-Int8": "Qwen/Qwen-1_8B-Chat-Int8",
+        "Llama-2-70b-chat-hf": "meta-llama/Llama-2-70b-chat-hf",
-        "Qwen-1_8B-Chat-Int4": "Qwen/Qwen-1_8B-Chat-Int4",
 
-        "Qwen-7B": "Qwen/Qwen-7B",
+        "Qwen-1_8B-Chat": "/media/checkpoint/Qwen-1_8B-Chat",
         "Qwen-7B-Chat": "Qwen/Qwen-7B-Chat",
-
-        "Qwen-14B": "Qwen/Qwen-14B",
         "Qwen-14B-Chat": "Qwen/Qwen-14B-Chat",
-
-        "Qwen-14B-Chat-Int8": "Qwen/Qwen-14B-Chat-Int8",
-        # 在新版的transformers下需要手动修改模型的config.json文件，在quantization_config字典中
-        # 增加`disable_exllama:true` 字段才能启动qwen的量化模型
-        "Qwen-14B-Chat-Int4": "Qwen/Qwen-14B-Chat-Int4",
-
-        "Qwen-72B": "Qwen/Qwen-72B",
         "Qwen-72B-Chat": "Qwen/Qwen-72B-Chat",
-        "Qwen-72B-Chat-Int8": "Qwen/Qwen-72B-Chat-Int8",
-        "Qwen-72B-Chat-Int4": "Qwen/Qwen-72B-Chat-Int4",
 
-        "baichuan2-13b": "baichuan-inc/Baichuan2-13B-Chat",
+        "baichuan-7b-chat": "baichuan-inc/Baichuan-7B-Chat",
-        "baichuan2-7b": "baichuan-inc/Baichuan2-7B-Chat",
-
-        "baichuan-7b": "baichuan-inc/Baichuan-7B",
-        "baichuan-13b": "baichuan-inc/Baichuan-13B",
         "baichuan-13b-chat": "baichuan-inc/Baichuan-13B-Chat",
-
+        "baichuan2-7b-chat": "baichuan-inc/Baichuan2-7B-Chat",
-        "aquila-7b": "BAAI/Aquila-7B",
+        "baichuan2-13b-chat": "baichuan-inc/Baichuan2-13B-Chat",
-        "aquilachat-7b": "BAAI/AquilaChat-7B",
 
         "internlm-7b": "internlm/internlm-7b",
         "internlm-chat-7b": "internlm/internlm-chat-7b",
+        "internlm2-chat-7b": "internlm/internlm2-chat-7b",
+        "internlm2-chat-20b": "internlm/internlm2-chat-20b",
+
+        "BlueLM-7B-Chat": "vivo-ai/BlueLM-7B-Chat",
+        "BlueLM-7B-Chat-32k": "vivo-ai/BlueLM-7B-Chat-32k",
+
+        "Yi-34B-Chat": "https://huggingface.co/01-ai/Yi-34B-Chat",
+
+        "agentlm-7b": "THUDM/agentlm-7b",
+        "agentlm-13b": "THUDM/agentlm-13b",
+        "agentlm-70b": "THUDM/agentlm-70b",
 
         "falcon-7b": "tiiuae/falcon-7b",
-        "falcon-40b": "tiiuae/falcon-40b",
+        "falcon-40b": "tiiuae/falcon-40,b",
         "falcon-rw-7b": "tiiuae/falcon-rw-7b",
 
+        "aquila-7b": "BAAI/Aquila-7B",
+        "aquilachat-7b": "BAAI/AquilaChat-7B",
+        "open_llama_13b": "openlm-research/open_llama_13b",
+        "vicuna-13b-v1.5": "lmsys/vicuna-13b-v1.5",
+        "koala": "young-geng/koala",
+        "mpt-7b": "mosaicml/mpt-7b",
+        "mpt-7b-storywriter": "mosaicml/mpt-7b-storywriter",
+        "mpt-30b": "mosaicml/mpt-30b",
+        "opt-66b": "facebook/opt-66b",
+        "opt-iml-max-30b": "facebook/opt-iml-max-30b",
         "gpt2": "gpt2",
         "gpt2-xl": "gpt2-xl",
-
         "gpt-j-6b": "EleutherAI/gpt-j-6b",
         "gpt4all-j": "nomic-ai/gpt4all-j",
         "gpt-neox-20b": "EleutherAI/gpt-neox-20b",
@@ -231,63 +218,50 @@ MODEL_PATH = {
         "oasst-sft-4-pythia-12b-epoch-3.5": "OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5",
         "dolly-v2-12b": "databricks/dolly-v2-12b",
         "stablelm-tuned-alpha-7b": "stabilityai/stablelm-tuned-alpha-7b",
-
-        "Llama-2-13b-hf": "meta-llama/Llama-2-13b-hf",
-        "Llama-2-70b-hf": "meta-llama/Llama-2-70b-hf",
-        "open_llama_13b": "openlm-research/open_llama_13b",
-        "vicuna-13b-v1.3": "lmsys/vicuna-13b-v1.3",
-        "koala": "young-geng/koala",
-
-        "mpt-7b": "mosaicml/mpt-7b",
-        "mpt-7b-storywriter": "mosaicml/mpt-7b-storywriter",
-        "mpt-30b": "mosaicml/mpt-30b",
-        "opt-66b": "facebook/opt-66b",
-        "opt-iml-max-30b": "facebook/opt-iml-max-30b",
-
-        "agentlm-7b": "THUDM/agentlm-7b",
-        "agentlm-13b": "THUDM/agentlm-13b",
-        "agentlm-70b": "THUDM/agentlm-70b",
-
-        "Yi-34B-Chat": "01-ai/Yi-34B-Chat",
     },
-    "reranker":{
+    "reranker": {
-        "bge-reranker-large":"BAAI/bge-reranker-large",
+        "bge-reranker-large": "BAAI/bge-reranker-large",
-        "bge-reranker-base":"BAAI/bge-reranker-base",
+        "bge-reranker-base": "BAAI/bge-reranker-base",
-        #TODO 增加在线reranker，如cohere
     }
 }
 
-
 # 通常情况下不需要更改以下内容
 
 # nltk 模型存储路径
 NLTK_DATA_PATH = os.path.join(os.path.dirname(os.path.dirname(__file__)), "nltk_data")
 
+# 使用VLLM可能导致模型推理能力下降，无法完成Agent任务
 VLLM_MODEL_DICT = {
-    "aquila-7b": "BAAI/Aquila-7B",
-    "aquilachat-7b": "BAAI/AquilaChat-7B",
-
-    "baichuan-7b": "baichuan-inc/Baichuan-7B",
-    "baichuan-13b": "baichuan-inc/Baichuan-13B",
-    "baichuan-13b-chat": "baichuan-inc/Baichuan-13B-Chat",
-
     "chatglm2-6b": "THUDM/chatglm2-6b",
     "chatglm2-6b-32k": "THUDM/chatglm2-6b-32k",
     "chatglm3-6b": "THUDM/chatglm3-6b",
     "chatglm3-6b-32k": "THUDM/chatglm3-6b-32k",
 
+    "Llama-2-7b-chat-hf": "meta-llama/Llama-2-7b-chat-hf",
+    "Llama-2-13b-chat-hf": "meta-llama/Llama-2-13b-chat-hf",
+    "Llama-2-70b-chat-hf": "meta-llama/Llama-2-70b-chat-hf",
+
+    "Qwen-1_8B-Chat": "Qwen/Qwen-1_8B-Chat",
+    "Qwen-7B-Chat": "Qwen/Qwen-7B-Chat",
+    "Qwen-14B-Chat": "Qwen/Qwen-14B-Chat",
+    "Qwen-72B-Chat": "Qwen/Qwen-72B-Chat",
+
+    "baichuan-7b-chat": "baichuan-inc/Baichuan-7B-Chat",
+    "baichuan-13b-chat": "baichuan-inc/Baichuan-13B-Chat",
+    "baichuan2-7b-chat": "baichuan-inc/Baichuan-7B-Chat",
+    "baichuan2-13b-chat": "baichuan-inc/Baichuan-13B-Chat",
+
     "BlueLM-7B-Chat": "vivo-ai/BlueLM-7B-Chat",
     "BlueLM-7B-Chat-32k": "vivo-ai/BlueLM-7B-Chat-32k",
 
-    # 注意：bloom系列的tokenizer与model是分离的，因此虽然vllm支持，但与fschat框架不兼容
-    # "bloom": "bigscience/bloom",
-    # "bloomz": "bigscience/bloomz",
-    # "bloomz-560m": "bigscience/bloomz-560m",
-    # "bloomz-7b1": "bigscience/bloomz-7b1",
-    # "bloomz-1b7": "bigscience/bloomz-1b7",
-
     "internlm-7b": "internlm/internlm-7b",
     "internlm-chat-7b": "internlm/internlm-chat-7b",
+    "internlm2-chat-7b": "internlm/Models/internlm2-chat-7b",
+    "internlm2-chat-20b": "internlm/Models/internlm2-chat-20b",
+
+    "aquila-7b": "BAAI/Aquila-7B",
+    "aquilachat-7b": "BAAI/AquilaChat-7B",
+
     "falcon-7b": "tiiuae/falcon-7b",
     "falcon-40b": "tiiuae/falcon-40b",
     "falcon-rw-7b": "tiiuae/falcon-rw-7b",
@@ -300,8 +274,6 @@ VLLM_MODEL_DICT = {
     "oasst-sft-4-pythia-12b-epoch-3.5": "OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5",
     "dolly-v2-12b": "databricks/dolly-v2-12b",
     "stablelm-tuned-alpha-7b": "stabilityai/stablelm-tuned-alpha-7b",
-    "Llama-2-13b-hf": "meta-llama/Llama-2-13b-hf",
-    "Llama-2-70b-hf": "meta-llama/Llama-2-70b-hf",
     "open_llama_13b": "openlm-research/open_llama_13b",
     "vicuna-13b-v1.3": "lmsys/vicuna-13b-v1.3",
     "koala": "young-geng/koala",
@@ -311,32 +283,8 @@ VLLM_MODEL_DICT = {
     "opt-66b": "facebook/opt-66b",
     "opt-iml-max-30b": "facebook/opt-iml-max-30b",
 
-    "Qwen-1_8B": "Qwen/Qwen-1_8B",
-    "Qwen-1_8B-Chat": "Qwen/Qwen-1_8B-Chat",
-    "Qwen-1_8B-Chat-Int8": "Qwen/Qwen-1_8B-Chat-Int8",
-    "Qwen-1_8B-Chat-Int4": "Qwen/Qwen-1_8B-Chat-Int4",
-
-    "Qwen-7B": "Qwen/Qwen-7B",
-    "Qwen-7B-Chat": "Qwen/Qwen-7B-Chat",
-
-    "Qwen-14B": "Qwen/Qwen-14B",
-    "Qwen-14B-Chat": "Qwen/Qwen-14B-Chat",
-    "Qwen-14B-Chat-Int8": "Qwen/Qwen-14B-Chat-Int8",
-    "Qwen-14B-Chat-Int4": "Qwen/Qwen-14B-Chat-Int4",
-
-    "Qwen-72B": "Qwen/Qwen-72B",
-    "Qwen-72B-Chat": "Qwen/Qwen-72B-Chat",
-    "Qwen-72B-Chat-Int8": "Qwen/Qwen-72B-Chat-Int8",
-    "Qwen-72B-Chat-Int4": "Qwen/Qwen-72B-Chat-Int4",
-
-    "agentlm-7b": "THUDM/agentlm-7b",
-    "agentlm-13b": "THUDM/agentlm-13b",
-    "agentlm-70b": "THUDM/agentlm-70b",
-
 }
 
-# 你认为支持Agent能力的模型，可以在这里添加，添加后不会出现可视化界面的警告
-# 经过我们测试，原生支持Agent的模型仅有以下几个
 SUPPORT_AGENT_MODEL = [
     "azure-api",
     "openai-api",
@@ -344,4 +292,6 @@ SUPPORT_AGENT_MODEL = [
     "Qwen",
     "chatglm3",
     "xinghuo-api",
+    "internlm2-chat-7b",
+    "internlm2-chat-20b"
 ]

server/agent/tools/weather_check.py

@@ -20,6 +20,6 @@ def weather(location: str, api_key: str):
 
 
 def weathercheck(location: str):
-    return weather(location, "S8vrB4U_-c5mvAMiK")
+    return weather(location, "your keys")
 class WeatherInput(BaseModel):
-    location: str = Field(description="City name,include city and county,like '厦门'")
+    location: str = Field(description="City name,include city and county")

server/model_workers/gemini.py

@@ -18,7 +18,7 @@ class GeminiWorker(ApiModelWorker):
             **kwargs,
     ):
         kwargs.update(model_names=model_names, controller_addr=controller_addr, worker_addr=worker_addr)
-        kwargs.setdefault("context_len", 4096) #TODO 16K模型需要改成16384
+        kwargs.setdefault("context_len", 4096)
         super().__init__(**kwargs)
 
     def create_gemini_messages(self,messages) -> json:
@@ -47,10 +47,10 @@ class GeminiWorker(ApiModelWorker):
         params.load_config(self.model_names[0])
         data = self.create_gemini_messages(messages=params.messages)
         generationConfig=dict(
-            temperature = params.temperature,
+            temperature=params.temperature,
-            topK = 1,
+            topK=1,
-            topP = 1,
+            topP=1,
-            maxOutputTokens = 4096,
+            maxOutputTokens=4096,
             stopSequences=[]
         )