From ab28ee58ab2fb49dfed2850c5556ca4e2c525918 Mon Sep 17 00:00:00 2001 From: zhayujie Date: Sat, 14 Feb 2026 16:49:44 +0800 Subject: [PATCH] feat: add doubao-2.0-code model and update README --- README.md | 52 +++- bridge/bridge.py | 3 + common/const.py | 22 +- config-template.json | 10 +- docs/agent.md | 12 +- models/bot_factory.py | 3 + models/doubao/__init__.py | 0 models/doubao/doubao_bot.py | 520 ++++++++++++++++++++++++++++++++ models/doubao/doubao_session.py | 51 ++++ 9 files changed, 643 insertions(+), 30 deletions(-) create mode 100644 models/doubao/__init__.py create mode 100644 models/doubao/doubao_bot.py create mode 100644 models/doubao/doubao_session.py diff --git a/README.md b/README.md index 71b4c1c..f14bb3a 100644 --- a/README.md +++ b/README.md @@ -18,7 +18,7 @@ - ✅ **长期记忆：** 自动将对话记忆持久化至本地文件和数据库中，包括全局记忆和天级记忆，支持关键词及向量检索 - ✅ **技能系统：** 实现了Skills创建和运行的引擎，内置多种技能，并支持通过自然语言对话完成自定义Skills开发 - ✅ **多模态消息：** 支持对文本、图片、语音、文件等多类型消息进行解析、处理、生成、发送等操作 -- ✅ **多模型接入：** 支持OpenAI, Claude, Gemini, DeepSeek, MiniMax、GLM、Qwen、Kimi等国内外主流模型厂商 +- ✅ **多模型接入：** 支持OpenAI, Claude, Gemini, DeepSeek, MiniMax、GLM、Qwen、Kimi、Doubao等国内外主流模型厂商 - ✅ **多端部署：** 支持运行在本地计算机或服务器，可集成到网页、飞书、钉钉、微信公众号、企业微信应用中使用 - ✅ **知识库：** 集成企业知识库能力，让Agent成为专属数字员工，基于[LinkAI](https://link-ai.tech)平台实现 @@ -90,7 +90,7 @@ bash <(curl -sS https://cdn.link-ai.tech/code/cow/run.sh) 项目支持国内外主流厂商的模型接口，可选模型及配置说明参考：[模型说明](#模型说明)。 -> 注：Agent模式下推荐使用以下模型，可根据效果及成本综合选择：GLM(glm-4.7)、MiniMAx(MiniMax-M2.1)、Qwen(qwen3-max)、Claude(claude-opus-4-6、claude-sonnet-4-5、claude-sonnet-4-0)、Gemini(gemini-3-flash-preview、gemini-3-pro-preview) +> 注：Agent模式下推荐使用以下模型，可根据效果及成本综合选择：MiniMax(MiniMax-M2.5)、GLM(glm-5)、Kimi(kimi-k2.5)、Qwen(qwen3-max)、Claude(claude-sonnet-4-5)、Gemini(gemini-3-flash-preview) 同时支持使用 **LinkAI平台** 接口，可灵活切换 OpenAI、Claude、Gemini、DeepSeek、Qwen、Kimi 等多种常用模型，并支持知识库、工作流、插件等Agent能力，参考 [接口文档](https://docs.link-ai.tech/platform/api)。 @@ -136,9 +136,11 @@ pip3 install -r requirements-optional.txt # config.json 文件内容示例 { "channel_type": "web", # 接入渠道类型，默认为web，支持修改为:feishu,dingtalk,wechatcom_app,terminal,wechatmp,wechatmp_service - "model": "MiniMax-M2.1", # 模型名称 + "model": "MiniMax-M2.5", # 模型名称 "minimax_api_key": "", # MiniMax API Key "zhipu_ai_api_key": "", # 智谱GLM API Key + "moonshot_api_key": "", # Kimi/Moonshot API Key + "ark_api_key": "", # 豆包(火山方舟) API Key "dashscope_api_key": "", # 百炼(通义千问)API Key "claude_api_key": "", # Claude API Key "claude_api_base": "https://api.anthropic.com/v1", # Claude API 地址，修改可接入三方代理平台 @@ -173,7 +175,7 @@ pip3 install -r requirements-optional.txt
2. 其他配置 -+ `model`: 模型名称，Agent模式下推荐使用 `glm-4.7`、`MiniMax-M2.1`、`qwen3-max`、`claude-opus-4-6`、`claude-sonnet-4-5`、`claude-sonnet-4-0`、`gemini-3-flash-preview`、`gemini-3-pro-preview`，全部模型名称参考[common/const.py](https://github.com/zhayujie/chatgpt-on-wechat/blob/master/common/const.py)文件 ++ `model`: 模型名称，Agent模式下推荐使用 `MiniMax-M2.5`、`glm-5`、`kimi-k2.5`、`qwen3-max`、`claude-sonnet-4-5`、`gemini-3-flash-preview`，全部模型名称参考[common/const.py](https://github.com/zhayujie/chatgpt-on-wechat/blob/master/common/const.py)文件 + `character_desc`：普通对话模式下的机器人系统提示词。在Agent模式下该配置不生效，由工作空间中的文件内容构成。 + `subscribe_msg`：订阅消息，公众号和企业微信channel中请填写，当被订阅时会自动回复， 可使用特殊占位符。目前支持的占位符有{trigger_prefix}，在程序中它会自动替换成bot的触发词。
@@ -309,24 +311,24 @@ volumes: ```json { - "model": "MiniMax-M2.1", + "model": "MiniMax-M2.5", "minimax_api_key": "" } ``` - - `model`: 可填写 `MiniMax-M2.1、MiniMax-M2.1-lightning、MiniMax-M2、abab6.5-chat` 等 + - `model`: 可填写 `MiniMax-M2.5、MiniMax-M2.1、MiniMax-M2.1-lightning、MiniMax-M2、abab6.5-chat` 等 - `minimax_api_key`：MiniMax平台的API-KEY，在 [控制台](https://platform.minimaxi.com/user-center/basic-information/interface-key) 创建 方式二：OpenAI兼容方式接入，配置如下： ```json { "bot_type": "chatGPT", - "model": "MiniMax-M2.1", + "model": "MiniMax-M2.5", "open_ai_api_base": "https://api.minimaxi.com/v1", "open_ai_api_key": "" } ``` - `bot_type`: OpenAI兼容方式 -- `model`: 可填 `MiniMax-M2.1、MiniMax-M2.1-lightning、MiniMax-M2`，参考[API文档](https://platform.minimaxi.com/document/%E5%AF%B9%E8%AF%9D?key=66701d281d57f38758d581d0#QklxsNSbaf6kM4j6wjO5eEek) +- `model`: 可填 `MiniMax-M2.5、MiniMax-M2.1、MiniMax-M2.1-lightning、MiniMax-M2`，参考[API文档](https://platform.minimaxi.com/document/%E5%AF%B9%E8%AF%9D?key=66701d281d57f38758d581d0#QklxsNSbaf6kM4j6wjO5eEek) - `open_ai_api_base`: MiniMax平台API的 BASE URL - `open_ai_api_key`: MiniMax平台的API-KEY @@ -338,24 +340,24 @@ volumes: ```json { - "model": "glm-4.7", + "model": "glm-5", "zhipu_ai_api_key": "" } ``` - - `model`: 可填 `glm-4.7、glm-4-plus、glm-4-flash、glm-4-air、glm-4-airx、glm-4-long` 等, 参考 [glm-4系列模型编码](https://bigmodel.cn/dev/api/normal-model/glm-4) + - `model`: 可填 `glm-5、glm-4.7、glm-4-plus、glm-4-flash、glm-4-air、glm-4-airx、glm-4-long` 等, 参考 [glm系列模型编码](https://bigmodel.cn/dev/api/normal-model/glm-4) - `zhipu_ai_api_key`: 智谱AI平台的 API KEY，在 [控制台](https://www.bigmodel.cn/usercenter/proj-mgmt/apikeys) 创建 方式二：OpenAI兼容方式接入，配置如下： ```json { "bot_type": "chatGPT", - "model": "glm-4.7", + "model": "glm-5", "open_ai_api_base": "https://open.bigmodel.cn/api/paas/v4", "open_ai_api_key": "" } ``` - `bot_type`: OpenAI兼容方式 -- `model`: 可填 `glm-4.7、glm-4.6、glm-4-plus、glm-4-flash、glm-4-air、glm-4-airx、glm-4-long` 等 +- `model`: 可填 `glm-5、glm-4.7、glm-4-plus、glm-4-flash、glm-4-air、glm-4-airx、glm-4-long` 等 - `open_ai_api_base`: 智谱AI平台的 BASE URL - `open_ai_api_key`: 智谱AI平台的 API KEY @@ -448,28 +450,46 @@ API Key创建：在 [控制台](https://aistudio.google.com/app/apikey?hl=zh-cn) ```json { - "model": "moonshot-v1-128k", + "model": "kimi-k2.5", "moonshot_api_key": "" } ``` - - `model`: 可填写 `moonshot-v1-8k、moonshot-v1-32k、moonshot-v1-128k` + - `model`: 可填写 `kimi-k2.5、kimi-k2、moonshot-v1-8k、moonshot-v1-32k、moonshot-v1-128k` - `moonshot_api_key`: Moonshot的API-KEY，在 [控制台](https://platform.moonshot.cn/console/api-keys) 创建 方式二：OpenAI兼容方式接入，配置如下： ```json { "bot_type": "chatGPT", - "model": "moonshot-v1-128k", + "model": "kimi-k2.5", "open_ai_api_base": "https://api.moonshot.cn/v1", "open_ai_api_key": "" } ``` - `bot_type`: OpenAI兼容方式 -- `model`: 可填写 `moonshot-v1-8k、moonshot-v1-32k、moonshot-v1-128k` +- `model`: 可填写 `kimi-k2.5、kimi-k2、moonshot-v1-8k、moonshot-v1-32k、moonshot-v1-128k` - `open_ai_api_base`: Moonshot的 BASE URL - `open_ai_api_key`: Moonshot的 API-KEY +
+豆包 (Doubao) + +1. API Key创建：在 [火山方舟控制台](https://console.volcengine.com/ark/region:ark+cn-beijing/apikey) 创建API Key + +2. 填写配置 + +```json +{ + "model": "doubao-seed-2-0-code-preview-260215", + "ark_api_key": "YOUR_API_KEY" +} +``` + - `model`: 可填写 `doubao-seed-2-0-code-preview-260215、doubao-seed-2-0-pro-260215、doubao-seed-2-0-lite-260215、doubao-seed-2-0-mini-260215` 等 + - `ark_api_key`: 火山方舟平台的 API Key，在 [控制台](https://console.volcengine.com/ark/region:ark+cn-beijing/apikey) 创建 + - `ark_base_url`: 可选，默认为 `https://ark.cn-beijing.volces.com/api/v3` +
+
Azure diff --git a/bridge/bridge.py b/bridge/bridge.py index 63fbb47..a4a5011 100644 --- a/bridge/bridge.py +++ b/bridge/bridge.py @@ -58,6 +58,9 @@ class Bridge(object): if model_type and model_type.startswith("kimi"): self.btype["chat"] = const.MOONSHOT + if model_type and model_type.startswith("doubao"): + self.btype["chat"] = const.DOUBAO + if model_type in [const.MODELSCOPE]: self.btype["chat"] = const.MODELSCOPE diff --git a/common/const.py b/common/const.py index bf93703..b44a081 100644 --- a/common/const.py +++ b/common/const.py @@ -83,12 +83,14 @@ QWEN3_MAX = "qwen3-max" # Qwen3 Max - Agent推荐模型 QWQ_PLUS = "qwq-plus" # MiniMax +MINIMAX_M2_5 = "MiniMax-M2.5" # MiniMax M2.5 - Latest MINIMAX_M2_1 = "MiniMax-M2.1" # MiniMax M2.1 - Agent推荐模型 MINIMAX_M2_1_LIGHTNING = "MiniMax-M2.1-lightning" # MiniMax M2.1 极速版 MINIMAX_M2 = "MiniMax-M2" # MiniMax M2 MINIMAX_ABAB6_5 = "abab6.5-chat" # MiniMax abab6.5 # GLM (智谱AI) +GLM_5 = "glm-5" # 智谱 GLM-5 - Latest GLM_4 = "glm-4" GLM_4_PLUS = "glm-4-plus" GLM_4_flash = "glm-4-flash" @@ -104,6 +106,13 @@ MOONSHOT = "moonshot" KIMI_K2 = "kimi-k2" KIMI_K2_5 = "kimi-k2.5" +# Doubao (Volcengine Ark) +DOUBAO = "doubao" +DOUBAO_SEED_2_CODE = "doubao-seed-2-0-code-preview-260215" +DOUBAO_SEED_2_PRO = "doubao-seed-2-0-pro-260215" +DOUBAO_SEED_2_LITE = "doubao-seed-2-0-lite-260215" +DOUBAO_SEED_2_MINI = "doubao-seed-2-0-mini-260215" + # 其他模型 WEN_XIN = "wenxin" WEN_XIN_4 = "wenxin-4" @@ -147,16 +156,19 @@ MODEL_LIST = [ QWEN, QWEN_TURBO, QWEN_PLUS, QWEN_MAX, QWEN_LONG, QWEN3_MAX, # MiniMax - MiniMax, MINIMAX_M2_1, MINIMAX_M2_1_LIGHTNING, MINIMAX_M2, MINIMAX_ABAB6_5, - + MiniMax, MINIMAX_M2_5, MINIMAX_M2_1, MINIMAX_M2_1_LIGHTNING, MINIMAX_M2, MINIMAX_ABAB6_5, + # GLM - ZHIPU_AI, GLM_4, GLM_4_PLUS, GLM_4_flash, GLM_4_LONG, GLM_4_ALLTOOLS, + ZHIPU_AI, GLM_5, GLM_4, GLM_4_PLUS, GLM_4_flash, GLM_4_LONG, GLM_4_ALLTOOLS, GLM_4_0520, GLM_4_AIR, GLM_4_AIRX, GLM_4_7, - + # Kimi MOONSHOT, "moonshot-v1-8k", "moonshot-v1-32k", "moonshot-v1-128k", KIMI_K2, KIMI_K2_5, - + + # Doubao + DOUBAO, DOUBAO_SEED_2_CODE, DOUBAO_SEED_2_PRO, DOUBAO_SEED_2_LITE, DOUBAO_SEED_2_MINI, + # 其他模型 WEN_XIN, WEN_XIN_4, XUNFEI, LINKAI_35, LINKAI_4_TURBO, LINKAI_4o, diff --git a/config-template.json b/config-template.json index d09d32a..d7cf86e 100644 --- a/config-template.json +++ b/config-template.json @@ -1,15 +1,17 @@ { "channel_type": "web", - "model": "glm-4.7", + "model": "MiniMax-M2.5", + "minimax_api_key": "", + "zhipu_ai_api_key": "", + "ark_api_key": "", + "moonshot_api_key": "", + "dashscope_api_key": "", "claude_api_key": "", "claude_api_base": "https://api.anthropic.com/v1", "open_ai_api_key": "", "open_ai_api_base": "https://api.openai.com/v1", "gemini_api_key": "", "gemini_api_base": "https://generativelanguage.googleapis.com", - "zhipu_ai_api_key": "", - "minimax_api_key": "", - "dashscope_api_key": "", "voice_to_text": "openai", "text_to_voice": "openai", "voice_reply_voice": false, diff --git a/docs/agent.md b/docs/agent.md index 1bf8a82..2e4e68f 100644 --- a/docs/agent.md +++ b/docs/agent.md @@ -8,7 +8,7 @@ Cow项目从简单的聊天机器人全面升级为超级智能助理 **CowAgent - **工具系统**：内置实现10+种工具，包括文件读写、bash终端、浏览器、定时任务、记忆管理等，通过Agent管理你的计算机或服务器 - **长期记忆**：自动将对话记忆持久化至本地文件和数据库中，包括全局记忆和天级记忆，支持关键词及向量检索 - **Skills系统**：新增Skill运行引擎，内置多种技能，并支持通过自然语言对话完成自定义Skills开发 -- **多渠道和多模型支持**：支持在Web、飞书、钉钉、企微等多渠道与Agent交互，支持Claude、Gemini、OpenAI、GLM、MiniMax、Qwen 等多种国内外主流模型 +- **多渠道和多模型支持**：支持在Web、飞书、钉钉、企微等多渠道与Agent交互，支持Claude、Gemini、OpenAI、GLM、MiniMax、Qwen、Kimi、Doubao 等多种国内外主流模型 - **安全和成本**：通过秘钥管理工具、提示词控制、系统权限等手段控制Agent的访问安全；通过最大记忆轮次、最大上下文token、工具执行步数对token成本进行限制 @@ -137,11 +137,13 @@ bash <(curl -sS https://cdn.link-ai.tech/code/cow/run.sh) Agent模式推荐使用以下模型，可根据效果及成本综合选择： -- **MiniMax**: `MiniMax-M2.1` -- **GLM**: `glm-4.7` +- **MiniMax**: `MiniMax-M2.5` +- **GLM**: `glm-5` +- **Kimi**: `kimi-k2.5` +- **Doubao**: `doubao-seed-2-0-code-preview-260215` - **Qwen**: `qwen3-max` -- **Claude**: `claude-sonnet-4-5`、`claude-sonnet-4-0` -- **Gemini**: `gemini-3-flash-preview`、`gemini-3-pro-preview` +- **Claude**: `claude-sonnet-4-5` +- **Gemini**: `gemini-3-flash-preview` 详细模型配置方式参考 [README.md 模型说明](../README.md#模型说明) diff --git a/models/bot_factory.py b/models/bot_factory.py index 3027d47..2f83da9 100644 --- a/models/bot_factory.py +++ b/models/bot_factory.py @@ -69,5 +69,8 @@ def create_bot(bot_type): from models.modelscope.modelscope_bot import ModelScopeBot return ModelScopeBot() + elif bot_type == const.DOUBAO: + from models.doubao.doubao_bot import DoubaoBot + return DoubaoBot() raise RuntimeError diff --git a/models/doubao/__init__.py b/models/doubao/__init__.py new file mode 100644 index 0000000..e69de29 diff --git a/models/doubao/doubao_bot.py b/models/doubao/doubao_bot.py new file mode 100644 index 0000000..987d718 --- /dev/null +++ b/models/doubao/doubao_bot.py @@ -0,0 +1,520 @@ +# encoding:utf-8 + +import json +import time + +import requests +from models.bot import Bot +from models.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 .doubao_session import DoubaoSession + + +# Doubao (火山方舟 / Volcengine Ark) API Bot +class DoubaoBot(Bot): + def __init__(self): + super().__init__() + self.sessions = SessionManager(DoubaoSession, model=conf().get("model") or "doubao-seed-2-0-pro-260215") + model = conf().get("model") or "doubao-seed-2-0-pro-260215" + self.args = { + "model": model, + "temperature": conf().get("temperature", 0.8), + "top_p": conf().get("top_p", 1.0), + } + self.api_key = conf().get("ark_api_key") + self.base_url = conf().get("ark_base_url", "https://ark.cn-beijing.volces.com/api/v3") + # Ensure base_url does not end with /chat/completions + if self.base_url.endswith("/chat/completions"): + self.base_url = self.base_url.rsplit("/chat/completions", 1)[0] + if self.base_url.endswith("/"): + self.base_url = self.base_url.rstrip("/") + + def reply(self, query, context=None): + # acquire reply content + if context.type == ContextType.TEXT: + logger.info("[DOUBAO] 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("[DOUBAO] session query={}".format(session.messages)) + + model = context.get("doubao_model") + new_args = self.args.copy() + if model: + new_args["model"] = model + + reply_content = self.reply_text(session, args=new_args) + logger.debug( + "[DOUBAO] 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("[DOUBAO] 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: DoubaoSession, args=None, retry_count: int = 0) -> dict: + """ + Call Doubao chat completion API to get the answer + :param session: a conversation session + :param args: model args + :param retry_count: retry count + :return: {} + """ + try: + headers = { + "Content-Type": "application/json", + "Authorization": "Bearer " + self.api_key + } + body = args.copy() + body["messages"] = session.messages + # Disable thinking by default for better efficiency + body["thinking"] = {"type": "disabled"} + res = requests.post( + f"{self.base_url}/chat/completions", + headers=headers, + json=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"[DOUBAO] 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: + logger.warn(f"[DOUBAO] 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 + + # ==================== Agent mode support ==================== + + def call_with_tools(self, messages, tools=None, stream: bool = False, **kwargs): + """ + Call Doubao API with tool support for agent integration. + + This method handles: + 1. Format conversion (Claude format -> OpenAI format) + 2. System prompt injection + 3. Streaming SSE response with tool_calls + 4. Thinking (reasoning) is disabled by default for efficiency + + Args: + messages: List of messages (may be in Claude format from agent) + tools: List of tool definitions (may be in Claude format from agent) + stream: Whether to use streaming + **kwargs: Additional parameters (max_tokens, temperature, system, model, etc.) + + Returns: + Generator yielding OpenAI-format chunks (for streaming) + """ + try: + # Convert messages from Claude format to OpenAI format + converted_messages = self._convert_messages_to_openai_format(messages) + + # Inject system prompt if provided + system_prompt = kwargs.pop("system", None) + if system_prompt: + if not converted_messages or converted_messages[0].get("role") != "system": + converted_messages.insert(0, {"role": "system", "content": system_prompt}) + else: + converted_messages[0] = {"role": "system", "content": system_prompt} + + # Convert tools from Claude format to OpenAI format + converted_tools = None + if tools: + converted_tools = self._convert_tools_to_openai_format(tools) + + # Resolve model / temperature + model = kwargs.pop("model", None) or self.args["model"] + max_tokens = kwargs.pop("max_tokens", None) + # Don't pop temperature, just ignore it - let API use default + kwargs.pop("temperature", None) + + # Build request body (omit temperature, let the API use its own default) + request_body = { + "model": model, + "messages": converted_messages, + "stream": stream, + } + if max_tokens is not None: + request_body["max_tokens"] = max_tokens + + # Add tools + if converted_tools: + request_body["tools"] = converted_tools + request_body["tool_choice"] = "auto" + + # Explicitly disable thinking to avoid reasoning_content issues + # in multi-turn tool calls + request_body["thinking"] = {"type": "disabled"} + + logger.debug(f"[DOUBAO] API call: model={model}, " + f"tools={len(converted_tools) if converted_tools else 0}, stream={stream}") + + if stream: + return self._handle_stream_response(request_body) + else: + return self._handle_sync_response(request_body) + + except Exception as e: + logger.error(f"[DOUBAO] call_with_tools error: {e}") + import traceback + logger.error(traceback.format_exc()) + + def error_generator(): + yield {"error": True, "message": str(e), "status_code": 500} + return error_generator() + + # -------------------- streaming -------------------- + + def _handle_stream_response(self, request_body: dict): + """Handle streaming SSE response from Doubao API and yield OpenAI-format chunks.""" + try: + headers = { + "Content-Type": "application/json", + "Authorization": f"Bearer {self.api_key}" + } + + url = f"{self.base_url}/chat/completions" + response = requests.post(url, headers=headers, json=request_body, stream=True, timeout=120) + + if response.status_code != 200: + error_msg = response.text + logger.error(f"[DOUBAO] API error: status={response.status_code}, msg={error_msg}") + yield {"error": True, "message": error_msg, "status_code": response.status_code} + return + + current_tool_calls = {} + finish_reason = None + + for line in response.iter_lines(): + if not line: + continue + + line = line.decode("utf-8") + if not line.startswith("data: "): + continue + + data_str = line[6:] # Remove "data: " prefix + if data_str.strip() == "[DONE]": + break + + try: + chunk = json.loads(data_str) + except json.JSONDecodeError as e: + logger.warning(f"[DOUBAO] JSON decode error: {e}, data: {data_str[:200]}") + continue + + # Check for error in chunk + if chunk.get("error"): + error_data = chunk["error"] + error_msg = error_data.get("message", "Unknown error") if isinstance(error_data, dict) else str(error_data) + logger.error(f"[DOUBAO] stream error: {error_msg}") + yield {"error": True, "message": error_msg, "status_code": 500} + return + + if not chunk.get("choices"): + continue + + choice = chunk["choices"][0] + delta = choice.get("delta", {}) + + # Skip reasoning_content (thinking) - don't log or forward + if delta.get("reasoning_content"): + continue + + # Handle text content + if "content" in delta and delta["content"]: + yield { + "choices": [{ + "index": 0, + "delta": { + "role": "assistant", + "content": delta["content"] + } + }] + } + + # Handle tool_calls (streamed incrementally) + if "tool_calls" in delta: + for tool_call_chunk in delta["tool_calls"]: + index = tool_call_chunk.get("index", 0) + if index not in current_tool_calls: + current_tool_calls[index] = { + "id": tool_call_chunk.get("id", ""), + "type": "tool_use", + "name": tool_call_chunk.get("function", {}).get("name", ""), + "input": "" + } + + # Accumulate arguments + if "function" in tool_call_chunk and "arguments" in tool_call_chunk["function"]: + current_tool_calls[index]["input"] += tool_call_chunk["function"]["arguments"] + + # Yield OpenAI-format tool call delta + yield { + "choices": [{ + "index": 0, + "delta": { + "tool_calls": [tool_call_chunk] + } + }] + } + + # Capture finish_reason + if choice.get("finish_reason"): + finish_reason = choice["finish_reason"] + + # Final chunk with finish_reason + yield { + "choices": [{ + "index": 0, + "delta": {}, + "finish_reason": finish_reason + }] + } + + except requests.exceptions.Timeout: + logger.error("[DOUBAO] Request timeout") + yield {"error": True, "message": "Request timeout", "status_code": 500} + except Exception as e: + logger.error(f"[DOUBAO] stream response error: {e}") + import traceback + logger.error(traceback.format_exc()) + yield {"error": True, "message": str(e), "status_code": 500} + + # -------------------- sync -------------------- + + def _handle_sync_response(self, request_body: dict): + """Handle synchronous API response and yield a single result dict.""" + try: + headers = { + "Content-Type": "application/json", + "Authorization": f"Bearer {self.api_key}" + } + + request_body.pop("stream", None) + url = f"{self.base_url}/chat/completions" + response = requests.post(url, headers=headers, json=request_body, timeout=120) + + if response.status_code != 200: + error_msg = response.text + logger.error(f"[DOUBAO] API error: status={response.status_code}, msg={error_msg}") + yield {"error": True, "message": error_msg, "status_code": response.status_code} + return + + result = response.json() + message = result["choices"][0]["message"] + finish_reason = result["choices"][0]["finish_reason"] + + response_data = {"role": "assistant", "content": []} + + # Add text content + if message.get("content"): + response_data["content"].append({ + "type": "text", + "text": message["content"] + }) + + # Add tool calls + if message.get("tool_calls"): + for tool_call in message["tool_calls"]: + response_data["content"].append({ + "type": "tool_use", + "id": tool_call["id"], + "name": tool_call["function"]["name"], + "input": json.loads(tool_call["function"]["arguments"]) + }) + + # Map finish_reason + if finish_reason == "tool_calls": + response_data["stop_reason"] = "tool_use" + elif finish_reason == "stop": + response_data["stop_reason"] = "end_turn" + else: + response_data["stop_reason"] = finish_reason + + yield response_data + + except requests.exceptions.Timeout: + logger.error("[DOUBAO] Request timeout") + yield {"error": True, "message": "Request timeout", "status_code": 500} + except Exception as e: + logger.error(f"[DOUBAO] sync response error: {e}") + import traceback + logger.error(traceback.format_exc()) + yield {"error": True, "message": str(e), "status_code": 500} + + # -------------------- format conversion -------------------- + + def _convert_messages_to_openai_format(self, messages): + """ + Convert messages from Claude format to OpenAI format. + + Claude format uses content blocks: tool_use / tool_result / text + OpenAI format uses tool_calls in assistant, role=tool for results + """ + if not messages: + return [] + + converted = [] + + for msg in messages: + role = msg.get("role") + content = msg.get("content") + + # Already a simple string - pass through + if isinstance(content, str): + converted.append(msg) + continue + + if not isinstance(content, list): + converted.append(msg) + continue + + if role == "user": + text_parts = [] + tool_results = [] + + for block in content: + if not isinstance(block, dict): + continue + if block.get("type") == "text": + text_parts.append(block.get("text", "")) + elif block.get("type") == "tool_result": + tool_call_id = block.get("tool_use_id") or "" + result_content = block.get("content", "") + if not isinstance(result_content, str): + result_content = json.dumps(result_content, ensure_ascii=False) + tool_results.append({ + "role": "tool", + "tool_call_id": tool_call_id, + "content": result_content + }) + + # Tool results first (must come right after assistant with tool_calls) + for tr in tool_results: + converted.append(tr) + + if text_parts: + converted.append({"role": "user", "content": "\n".join(text_parts)}) + + elif role == "assistant": + openai_msg = {"role": "assistant"} + text_parts = [] + tool_calls = [] + + for block in content: + if not isinstance(block, dict): + continue + if block.get("type") == "text": + text_parts.append(block.get("text", "")) + elif block.get("type") == "tool_use": + tool_calls.append({ + "id": block.get("id"), + "type": "function", + "function": { + "name": block.get("name"), + "arguments": json.dumps(block.get("input", {})) + } + }) + + if text_parts: + openai_msg["content"] = "\n".join(text_parts) + elif not tool_calls: + openai_msg["content"] = "" + + if tool_calls: + openai_msg["tool_calls"] = tool_calls + if not text_parts: + openai_msg["content"] = None + + converted.append(openai_msg) + else: + converted.append(msg) + + return converted + + def _convert_tools_to_openai_format(self, tools): + """ + Convert tools from Claude format to OpenAI format. + + Claude: {name, description, input_schema} + OpenAI: {type: "function", function: {name, description, parameters}} + """ + if not tools: + return None + + converted = [] + for tool in tools: + # Already in OpenAI format + if "type" in tool and tool["type"] == "function": + converted.append(tool) + else: + converted.append({ + "type": "function", + "function": { + "name": tool.get("name"), + "description": tool.get("description"), + "parameters": tool.get("input_schema", {}) + } + }) + + return converted diff --git a/models/doubao/doubao_session.py b/models/doubao/doubao_session.py new file mode 100644 index 0000000..561347e --- /dev/null +++ b/models/doubao/doubao_session.py @@ -0,0 +1,51 @@ +from models.session_manager import Session +from common.log import logger + + +class DoubaoSession(Session): + def __init__(self, session_id, system_prompt=None, model="doubao-seed-2-0-pro-260215"): + 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