diff --git a/bot/bot_factory.py b/bot/bot_factory.py index a6ef241..05cf0ea 100644 --- a/bot/bot_factory.py +++ b/bot/bot_factory.py @@ -68,5 +68,9 @@ def create_bot(bot_type): from bot.minimax.minimax_bot import MinimaxBot return MinimaxBot() + elif bot_type == const.MODELSCOPE: + from bot.modelscope.modelscope_bot import ModelScopeBot + return ModelScopeBot() + raise RuntimeError diff --git a/bot/modelscope/modelscope_bot.py b/bot/modelscope/modelscope_bot.py new file mode 100644 index 0000000..159d425 --- /dev/null +++ b/bot/modelscope/modelscope_bot.py @@ -0,0 +1,152 @@ +# encoding:utf-8 + +import time +import json +import openai +import openai.error +from bot.bot import Bot +from bot.session_manager import SessionManager +from bridge.context import ContextType +from bridge.reply import Reply, ReplyType +from common.log import logger +from config import conf, load_config +from .modelscope_session import ModelScopeSession +import requests + + +# ModelScope对话模型API +class ModelScopeBot(Bot): + def __init__(self): + super().__init__() + self.sessions = SessionManager(ModelScopeSession, model=conf().get("model") or "Qwen/Qwen2.5-7B-Instruct") + model = conf().get("model") or "Qwen/Qwen2.5-7B-Instruct" + if model == "modelscope": + model = "Qwen/Qwen2.5-7B-Instruct" + self.args = { + "model": model, # 对话模型的名称 + "temperature": conf().get("temperature", 0.3), # 如果设置，值域须为 [0, 1] 我们推荐 0.3，以达到较合适的效果。 + "top_p": conf().get("top_p", 1.0), # 使用默认值 + } + self.api_key = conf().get("modelscope_api_key") + self.base_url = conf().get("modelscope_base_url", "https://api-inference.modelscope.cn/v1/chat/completions") + """ + 需要获取MOdelScope支持API-inference的模型名称列表，请到魔搭社区官网模型中心查看 https://modelscope.cn/models?filter=inference_type&page=1。 + 或者使用命令 curl https://api-inference.modelscope.cn/v1/models 对模型列表和ID进行获取。 + 获取ModelScope的免费API Key，请到魔搭社区官网用户中心查看获取方式 https://modelscope.cn/docs/model-service/API-Inference/intro。 + """ + def reply(self, query, context=None): + # acquire reply content + if context.type == ContextType.TEXT: + logger.info("[MODELSCOPE_AI] query={}".format(query)) + + session_id = context["session_id"] + reply = None + clear_memory_commands = conf().get("clear_memory_commands", ["#清除记忆"]) + if query in clear_memory_commands: + self.sessions.clear_session(session_id) + reply = Reply(ReplyType.INFO, "记忆已清除") + elif query == "#清除所有": + self.sessions.clear_all_session() + reply = Reply(ReplyType.INFO, "所有人记忆已清除") + elif query == "#更新配置": + load_config() + reply = Reply(ReplyType.INFO, "配置已更新") + if reply: + return reply + session = self.sessions.session_query(query, session_id) + logger.debug("[MODELSCOPE_AI] session query={}".format(session.messages)) + + model = context.get("modelscope_model") + new_args = self.args.copy() + if model: + new_args["model"] = model + # if context.get('stream'): + # # reply in stream + # return self.reply_text_stream(query, new_query, session_id) + + reply_content = self.reply_text(session, args=new_args) + logger.debug( + "[MODELSCOPE_AI] new_query={}, session_id={}, reply_cont={}, completion_tokens={}".format( + session.messages, + session_id, + reply_content["content"], + reply_content["completion_tokens"], + ) + ) + if reply_content["completion_tokens"] == 0 and len(reply_content["content"]) > 0: + reply = Reply(ReplyType.ERROR, reply_content["content"]) + elif reply_content["completion_tokens"] > 0: + self.sessions.session_reply(reply_content["content"], session_id, reply_content["total_tokens"]) + reply = Reply(ReplyType.TEXT, reply_content["content"]) + else: + reply = Reply(ReplyType.ERROR, reply_content["content"]) + logger.debug("[MODELSCOPE_AI] reply {} used 0 tokens.".format(reply_content)) + return reply + else: + reply = Reply(ReplyType.ERROR, "Bot不支持处理{}类型的消息".format(context.type)) + return reply + + def reply_text(self, session: ModelScopeSession, args=None, retry_count=0) -> dict: + """ + call openai's ChatCompletion to get the answer + :param session: a conversation session + :param session_id: session id + :param retry_count: retry count + :return: {} + """ + try: + headers = { + "Content-Type": "application/json", + "Authorization": "Bearer " + self.api_key + } + + body = args + body["messages"] = session.messages + # logger.debug("[MODELSCOPE_AI] response={}".format(response)) + # logger.info("[MODELSCOPE_AI] reply={}, total_tokens={}".format(response.choices[0]['message']['content'], response["usage"]["total_tokens"])) + res = requests.post( + self.base_url, + headers=headers, + data=json.dumps(body) + ) + + if res.status_code == 200: + response = res.json() + return { + "total_tokens": response["usage"]["total_tokens"], + "completion_tokens": response["usage"]["completion_tokens"], + "content": response["choices"][0]["message"]["content"] + } + else: + response = res.json() + error = response.get("error") + logger.error(f"[MODELSCOPE_AI] chat failed, status_code={res.status_code}, " + f"msg={error.get('message')}, type={error.get('type')}") + + result = {"completion_tokens": 0, "content": "提问太快啦，请休息一下再问我吧"} + need_retry = False + if res.status_code >= 500: + # server error, need retry + logger.warn(f"[MODELSCOPE_AI] do retry, times={retry_count}") + need_retry = retry_count < 2 + elif res.status_code == 401: + result["content"] = "授权失败，请检查API Key是否正确" + elif res.status_code == 429: + result["content"] = "请求过于频繁，请稍后再试" + need_retry = retry_count < 2 + else: + need_retry = False + + if need_retry: + time.sleep(3) + return self.reply_text(session, args, retry_count + 1) + else: + return result + except Exception as e: + logger.exception(e) + need_retry = retry_count < 2 + result = {"completion_tokens": 0, "content": "我现在有点累了，等会再来吧"} + if need_retry: + return self.reply_text(session, args, retry_count + 1) + else: + return result diff --git a/bot/modelscope/modelscope_session.py b/bot/modelscope/modelscope_session.py new file mode 100644 index 0000000..726d683 --- /dev/null +++ b/bot/modelscope/modelscope_session.py @@ -0,0 +1,51 @@ +from bot.session_manager import Session +from common.log import logger + + +class ModelScopeSession(Session): + def __init__(self, session_id, system_prompt=None, model="Qwen/Qwen2.5-7B-Instruct"): + super().__init__(session_id, system_prompt) + self.model = model + self.reset() + + def discard_exceeding(self, max_tokens, cur_tokens=None): + precise = True + try: + cur_tokens = self.calc_tokens() + except Exception as e: + precise = False + if cur_tokens is None: + raise e + logger.debug("Exception when counting tokens precisely for query: {}".format(e)) + while cur_tokens > max_tokens: + if len(self.messages) > 2: + self.messages.pop(1) + elif len(self.messages) == 2 and self.messages[1]["role"] == "assistant": + self.messages.pop(1) + if precise: + cur_tokens = self.calc_tokens() + else: + cur_tokens = cur_tokens - max_tokens + break + elif len(self.messages) == 2 and self.messages[1]["role"] == "user": + logger.warn("user message exceed max_tokens. total_tokens={}".format(cur_tokens)) + break + else: + logger.debug("max_tokens={}, total_tokens={}, len(messages)={}".format(max_tokens, cur_tokens, + len(self.messages))) + break + if precise: + cur_tokens = self.calc_tokens() + else: + cur_tokens = cur_tokens - max_tokens + return cur_tokens + + def calc_tokens(self): + return num_tokens_from_messages(self.messages, self.model) + + +def num_tokens_from_messages(messages, model): + tokens = 0 + for msg in messages: + tokens += len(msg["content"]) + return tokens diff --git a/bridge/bridge.py b/bridge/bridge.py index 3e714a6..c50779d 100644 --- a/bridge/bridge.py +++ b/bridge/bridge.py @@ -49,6 +49,9 @@ class Bridge(object): if model_type in [const.MOONSHOT, "moonshot-v1-8k", "moonshot-v1-32k", "moonshot-v1-128k"]: self.btype["chat"] = const.MOONSHOT + if model_type in [const.MODELSCOPE,]: + self.btype["chat"] = const.MODELSCOPE + if model_type in ["abab6.5-chat"]: self.btype["chat"] = const.MiniMax diff --git a/common/const.py b/common/const.py index 9abbc41..4220ecb 100644 --- a/common/const.py +++ b/common/const.py @@ -15,7 +15,7 @@ GEMINI = "gemini" # gemini-1.0-pro ZHIPU_AI = "glm-4" MOONSHOT = "moonshot" MiniMax = "minimax" - +MODELSCOPE = "Qwen/Qwen2.5-7B-Instruct" # model CLAUDE3 = "claude-3-opus-20240229" @@ -97,7 +97,8 @@ MODEL_LIST = [ "moonshot-v1-8k", "moonshot-v1-32k", "moonshot-v1-128k", QWEN, QWEN_TURBO, QWEN_PLUS, QWEN_MAX, LINKAI_35, LINKAI_4_TURBO, LINKAI_4o, - DEEPSEEK_CHAT, DEEPSEEK_REASONER + DEEPSEEK_CHAT, DEEPSEEK_REASONER, + MODELSCOPE ] MODEL_LIST = MODEL_LIST + GITEE_AI_MODEL_LIST diff --git a/config.py b/config.py index c675089..a02bfae 100644 --- a/config.py +++ b/config.py @@ -171,6 +171,9 @@ available_setting = { "zhipu_ai_api_base": "https://open.bigmodel.cn/api/paas/v4", "moonshot_api_key": "", "moonshot_base_url": "https://api.moonshot.cn/v1/chat/completions", + #魔搭社区 平台配置 + "modelscope_api_key": "", + "modelscope_base_url": "https://api-inference.modelscope.cn/v1/chat/completions", # LinkAI平台配置 "use_linkai": False, "linkai_api_key": "", diff --git a/plugins/godcmd/godcmd.py b/plugins/godcmd/godcmd.py index f2889b3..fe35879 100644 --- a/plugins/godcmd/godcmd.py +++ b/plugins/godcmd/godcmd.py @@ -339,7 +339,8 @@ class Godcmd(Plugin): ok, result = True, "配置已重载" elif cmd == "resetall": if bottype in [const.OPEN_AI, const.CHATGPT, const.CHATGPTONAZURE, const.LINKAI, - const.BAIDU, const.XUNFEI, const.QWEN, const.GEMINI, const.ZHIPU_AI, const.MOONSHOT]: + const.BAIDU, const.XUNFEI, const.QWEN, const.GEMINI, const.ZHIPU_AI, const.MOONSHOT, + const.MODELSCOPE]: channel.cancel_all_session() bot.sessions.clear_all_session() ok, result = True, "重置所有会话成功" diff --git a/plugins/role/role.py b/plugins/role/role.py index 8890a62..87f5a14 100644 --- a/plugins/role/role.py +++ b/plugins/role/role.py @@ -99,7 +99,7 @@ class Role(Plugin): if e_context["context"].type != ContextType.TEXT: return btype = Bridge().get_bot_type("chat") - if btype not in [const.OPEN_AI, const.CHATGPT, const.CHATGPTONAZURE, const.QWEN_DASHSCOPE, const.XUNFEI, const.BAIDU, const.ZHIPU_AI, const.MOONSHOT, const.MiniMax, const.LINKAI]: + if btype not in [const.OPEN_AI, const.CHATGPT, const.CHATGPTONAZURE, const.QWEN_DASHSCOPE, const.XUNFEI, const.BAIDU, const.ZHIPU_AI, const.MOONSHOT, const.MiniMax, const.LINKAI,const.MODELSCOPE]: logger.debug(f'不支持的bot: {btype}') return bot = Bridge().get_bot("chat")