fix: Non streaming sessions cannot count token consumption #3635

.github/dependabot.yml

@@ -0,0 +1,17 @@
+version: 2
+updates:
+  - package-ecosystem: "pip"
+    directory: "/"
+    schedule:
+      interval: "weekly"
+      timezone: "Asia/Shanghai"
+      day: "friday"
+    target-branch: "v2"
+    groups:
+      python-dependencies:
+        patterns:
+          - "*"
+#    ignore:
+#      - dependency-name: "pymupdf"
+#        versions: ["*"]
+

.github/workflows/build-and-push.yml

@@ -7,7 +7,7 @@ on:
     inputs:
       dockerImageTag:
         description: 'Image Tag'
-        default: 'v1.10.3-dev'
+        default: 'v1.10.7-dev'
         required: true
       dockerImageTagWithLatest:
         description: '是否发布latest tag（正式发版时选择，测试版本切勿选择）'
@@ -36,7 +36,7 @@ on:
 jobs:
   build-and-push-to-fit2cloud-registry:
     if: ${{ contains(github.event.inputs.registry, 'fit2cloud') }}
-    runs-on: ubuntu-22.04
+    runs-on: ubuntu-latest
     steps:
       - name: Check Disk Space
         run: df -h
@@ -52,10 +52,6 @@ jobs:
           swap-storage: true
       - name: Check Disk Space
         run: df -h
-      - name: Set Swap Space
-        uses: pierotofy/set-swap-space@master
-        with:
-          swap-size-gb: 8
       - name: Checkout
         uses: actions/checkout@v4
         with:
@@ -68,24 +64,17 @@ jobs:
           TAG_NAME=${{ github.event.inputs.dockerImageTag }}
           TAG_NAME_WITH_LATEST=${{ github.event.inputs.dockerImageTagWithLatest }}
           if [[ ${TAG_NAME_WITH_LATEST} == 'true' ]]; then            
-            DOCKER_IMAGE_TAGS="--tag ${DOCKER_IMAGE}:${TAG_NAME} --tag ${DOCKER_IMAGE}:latest"
+            DOCKER_IMAGE_TAGS="--tag ${DOCKER_IMAGE}:${TAG_NAME} --tag ${DOCKER_IMAGE}:${TAG_NAME%%.*}"
           else
             DOCKER_IMAGE_TAGS="--tag ${DOCKER_IMAGE}:${TAG_NAME}"
           fi
           echo ::set-output name=buildx_args::--platform ${DOCKER_PLATFORMS} --memory-swap -1 \
-            --build-arg DOCKER_IMAGE_TAG=${{ github.event.inputs.dockerImageTag }} --build-arg BUILD_AT=$(TZ=Asia/Shanghai date +'%Y-%m-%dT%H:%M') --build-arg GITHUB_COMMIT=${GITHUB_SHA::8} --no-cache \
+            --build-arg DOCKER_IMAGE_TAG=${{ github.event.inputs.dockerImageTag }} --build-arg BUILD_AT=$(TZ=Asia/Shanghai date +'%Y-%m-%dT%H:%M') --build-arg GITHUB_COMMIT=`git rev-parse --short HEAD` --no-cache \
             ${DOCKER_IMAGE_TAGS} .
       - name: Set up QEMU
         uses: docker/setup-qemu-action@v3
-        with:
-          # Until https://github.com/tonistiigi/binfmt/issues/215
-          image: tonistiigi/binfmt:qemu-v7.0.0-28
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
-        with:
-          buildkitd-config-inline: |
-            [worker.oci]
-              max-parallelism = 1
       - name: Login to GitHub Container Registry
         uses: docker/login-action@v3
         with:
@@ -100,11 +89,12 @@ jobs:
           password: ${{ secrets.FIT2CLOUD_REGISTRY_PASSWORD }}
       - name: Docker Buildx (build-and-push)
         run: |
+          sudo sync && echo 3 | sudo tee /proc/sys/vm/drop_caches && free -m
           docker buildx build --output "type=image,push=true" ${{ steps.prepare.outputs.buildx_args }} -f installer/Dockerfile
 
   build-and-push-to-dockerhub:
     if: ${{ contains(github.event.inputs.registry, 'dockerhub') }}
-    runs-on: ubuntu-22.04
+    runs-on: ubuntu-latest
     steps:
       - name: Check Disk Space
         run: df -h
@@ -120,10 +110,6 @@ jobs:
           swap-storage: true
       - name: Check Disk Space
         run: df -h
-      - name: Set Swap Space
-        uses: pierotofy/set-swap-space@master
-        with:
-          swap-size-gb: 8
       - name: Checkout
         uses: actions/checkout@v4
         with:
@@ -136,24 +122,17 @@ jobs:
           TAG_NAME=${{ github.event.inputs.dockerImageTag }}          
           TAG_NAME_WITH_LATEST=${{ github.event.inputs.dockerImageTagWithLatest }}
           if [[ ${TAG_NAME_WITH_LATEST} == 'true' ]]; then            
-            DOCKER_IMAGE_TAGS="--tag ${DOCKER_IMAGE}:${TAG_NAME} --tag ${DOCKER_IMAGE}:latest"
+            DOCKER_IMAGE_TAGS="--tag ${DOCKER_IMAGE}:${TAG_NAME} --tag ${DOCKER_IMAGE}:${TAG_NAME%%.*}"
           else
             DOCKER_IMAGE_TAGS="--tag ${DOCKER_IMAGE}:${TAG_NAME}"
           fi
           echo ::set-output name=buildx_args::--platform ${DOCKER_PLATFORMS} --memory-swap -1 \
-            --build-arg DOCKER_IMAGE_TAG=${{ github.event.inputs.dockerImageTag }} --build-arg BUILD_AT=$(TZ=Asia/Shanghai date +'%Y-%m-%dT%H:%M') --build-arg GITHUB_COMMIT=${GITHUB_SHA::8} --no-cache \
+            --build-arg DOCKER_IMAGE_TAG=${{ github.event.inputs.dockerImageTag }} --build-arg BUILD_AT=$(TZ=Asia/Shanghai date +'%Y-%m-%dT%H:%M') --build-arg GITHUB_COMMIT=`git rev-parse --short HEAD` --no-cache \
             ${DOCKER_IMAGE_TAGS} .
       - name: Set up QEMU
         uses: docker/setup-qemu-action@v3
-        with:
-          # Until https://github.com/tonistiigi/binfmt/issues/215
-          image: tonistiigi/binfmt:qemu-v7.0.0-28
       - name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v3
-        with:
-          buildkitd-config-inline: |
-            [worker.oci]
-              max-parallelism = 1
       - name: Login to GitHub Container Registry
         uses: docker/login-action@v3
         with:
@@ -167,4 +146,5 @@ jobs:
           password: ${{ secrets.DOCKERHUB_TOKEN }}
       - name: Docker Buildx (build-and-push)
         run: |
+          sudo sync && echo 3 | sudo tee /proc/sys/vm/drop_caches && free -m
           docker buildx build --output "type=image,push=true" ${{ steps.prepare.outputs.buildx_args }} -f installer/Dockerfile

README.md

@@ -1,5 +1,6 @@
 <p align="center"><img src= "https://github.com/1Panel-dev/maxkb/assets/52996290/c0694996-0eed-40d8-b369-322bf2a380bf" alt="MaxKB" width="300" /></p>
-<h3 align="center">Ready-to-use AI Chatbot</h3>
+<h3 align="center">Open-source platform for building enterprise-grade agents</h3>
+<h3 align="center">强大易用的企业级智能体平台</h3>
 <p align="center"><a href="https://trendshift.io/repositories/9113" target="_blank"><img src="https://trendshift.io/api/badge/repositories/9113" alt="1Panel-dev%2FMaxKB | Trendshift" style="width: 250px; height: 55px;" width="250" height="55"/></a></p>
 <p align="center">
   <a href="https://www.gnu.org/licenses/gpl-3.0.html#license-text"><img src="https://img.shields.io/github/license/1Panel-dev/maxkb?color=%231890FF" alt="License: GPL v3"></a>
@@ -10,10 +11,10 @@
 </p>
 <hr/>
 
-MaxKB = Max Knowledge Base, it is a ready-to-use AI chatbot that integrates Retrieval-Augmented Generation (RAG) pipelines, supports robust workflows, and provides advanced MCP tool-use capabilities. MaxKB is widely applied in scenarios such as intelligent customer service, corporate internal knowledge bases, academic research, and education.
+MaxKB = Max Knowledge Brain, it is an open-source platform for building enterprise-grade agents. MaxKB integrates Retrieval-Augmented Generation (RAG) pipelines, supports robust workflows, and provides advanced MCP tool-use capabilities. MaxKB is widely applied in scenarios such as intelligent customer service, corporate internal knowledge bases, academic research, and education.
 
-- **RAG Pipeline**: Supports direct uploading of documents / automatic crawling of online documents, with features for automatic text splitting, vectorization, and RAG (Retrieval-Augmented Generation). This effectively reduces hallucinations in large models, providing a superior smart Q&A interaction experience.
-- **Flexible Orchestration**: Equipped with a powerful workflow engine, function library and MCP tool-use, enabling the orchestration of AI processes to meet the needs of complex business scenarios. 
+- **RAG Pipeline**: Supports direct uploading of documents / automatic crawling of online documents, with features for automatic text splitting, vectorization. This effectively reduces hallucinations in large models, providing a superior smart Q&A interaction experience.
+- **Agentic Workflow**: Equipped with a powerful workflow engine, function library and MCP tool-use, enabling the orchestration of AI processes to meet the needs of complex business scenarios. 
 - **Seamless Integration**: Facilitates zero-coding rapid integration into third-party business systems, quickly equipping existing systems with intelligent Q&A capabilities to enhance user satisfaction.
 - **Model-Agnostic**: Supports various large models, including private models (such as DeepSeek, Llama, Qwen, etc.) and public models (like OpenAI, Claude, Gemini, etc.).
 - **Multi Modal**: Native support for input and output text, image, audio and video.
@@ -23,15 +24,15 @@ MaxKB = Max Knowledge Base, it is a ready-to-use AI chatbot that integrates Retr
 Execute the script below to start a MaxKB container using Docker:
 
 ```bash
-docker run -d --name=maxkb --restart=always -p 8080:8080 -v ~/.maxkb:/var/lib/postgresql/data -v ~/.python-packages:/opt/maxkb/app/sandbox/python-packages 1panel/maxkb
+docker run -d --name=maxkb --restart=always -p 8080:8080 -v ~/.maxkb:/var/lib/postgresql/data -v ~/.python-packages:/opt/maxkb/app/sandbox/python-packages 1panel/maxkb:v1
 ```
 
 Access MaxKB web interface at `http://your_server_ip:8080` with default admin credentials:
 
 - username: admin
 - password: MaxKB@123..
 
-中国用户如遇到 Docker 镜像 Pull 失败问题，请参照该 [离线安装文档](https://maxkb.cn/docs/installation/offline_installtion/) 进行安装。
+中国用户如遇到 Docker 镜像 Pull 失败问题，请参照该 [离线安装文档](https://maxkb.cn/docs/v1/installation/offline_installtion/) 进行安装。
 
 ## Screenshots
 
@@ -55,8 +56,6 @@ Access MaxKB web interface at `http://your_server_ip:8080` with default admin cr
 
 ## Feature Comparison
 
-MaxKB is positioned as an Ready-to-use RAG (Retrieval-Augmented Generation) intelligent Q&A application, rather than a middleware platform for building large model applications. The following table is merely a comparison from a functional perspective.
-
 <table style="width: 100%;">
   <tr>
     <th align="center">Feature</th>

README_CN.md

@@ -1,45 +1,45 @@
 <p align="center"><img src= "https://github.com/1Panel-dev/maxkb/assets/52996290/c0694996-0eed-40d8-b369-322bf2a380bf" alt="MaxKB" width="300" /></p>
-<h3 align="center">基于大模型和 RAG 的知识库问答系统</h3>
-<h4 align="center">Ready-to-use, flexible RAG Chatbot</h4>
+<h3 align="center">强大易用的企业级智能体平台</h3>
 <p align="center">
     <a href="https://trendshift.io/repositories/9113" target="_blank"><img src="https://trendshift.io/api/badge/repositories/9113" alt="1Panel-dev%2FMaxKB | Trendshift" style="width: 250px; height: auto;" /></a>
-    <a href="https://market.aliyun.com/products/53690006/cmjj00067609.html?userCode=kmemb8jp" target="_blank"><img src="https://img.alicdn.com/imgextra/i2/O1CN01H5JIwY1rZ0OobDjnJ_!!6000000005644-2-tps-1000-216.png" alt="1Panel-dev%2FMaxKB | Aliyun" style="width: 250px; height: auto;" /></a>
 </p>
 <p align="center">
   <a href="README_EN.md"><img src="https://img.shields.io/badge/English_README-blue" alt="English README"></a>
-  <a href="https://www.gnu.org/licenses/gpl-3.0.html#license-text"><img src="https://img.shields.io/github/license/1Panel-dev/maxkb" alt="License: GPL v3"></a>
+  <a href="https://www.gnu.org/licenses/gpl-3.0.html#license-text"><img src="https://img.shields.io/github/license/1Panel-dev/maxkb?color=%231890FF" alt="License: GPL v3"></a>
   <a href="https://github.com/1Panel-dev/maxkb/releases/latest"><img src="https://img.shields.io/github/v/release/1Panel-dev/maxkb" alt="Latest release"></a>
-  <a href="https://github.com/1Panel-dev/maxkb"><img src="https://img.shields.io/github/stars/1Panel-dev/maxkb?style=flat-square" alt="Stars"></a>    
-  <a href="https://hub.docker.com/r/1panel/maxkb"><img src="https://img.shields.io/docker/pulls/1panel/maxkb?label=downloads" alt="Download"></a>  
+  <a href="https://github.com/1Panel-dev/maxkb"><img src="https://img.shields.io/github/stars/1Panel-dev/maxkb?style=flat-square" alt="Stars"></a>
+  <a href="https://hub.docker.com/r/1panel/maxkb"><img src="https://img.shields.io/docker/pulls/1panel/maxkb?label=downloads" alt="Download"></a>
+  <a href="https://gitee.com/fit2cloud-feizhiyun/MaxKB"><img src="https://gitee.com/fit2cloud-feizhiyun/MaxKB/badge/star.svg?theme=gvp" alt="Gitee Stars"></a>
+  <a href="https://gitcode.com/feizhiyun/MaxKB"><img src="https://gitcode.com/feizhiyun/MaxKB/star/badge.svg" alt="GitCode Stars"></a>
 </p>
 <hr/>
 
-MaxKB = Max Knowledge Base，是一款开箱即用的 RAG Chatbot，具备强大的工作流和 MCP 工具调用能力。它支持对接各种主流大语言模型（LLMs），广泛应用于智能客服、企业内部知识库、学术研究与教育等场景。
+MaxKB = Max Knowledge Brain，是一个强大易用的企业级智能体平台，致力于解决企业 AI 落地面临的技术门槛高、部署成本高、迭代周期长等问题，助力企业在人工智能时代赢得先机。秉承“开箱即用，伴随成长”的设计理念，MaxKB 支持企业快速接入主流大模型，高效构建专属知识库，并提供从基础问答（RAG）、复杂流程自动化（工作流）到智能体（Agent）的渐进式升级路径，全面赋能智能客服、智能办公助手等多种应用场景。
 
-- **开箱即用**：支持直接上传文档 / 自动爬取在线文档，支持文本自动拆分、向量化和 RAG（检索增强生成），有效减少大模型幻觉，智能问答交互体验好；
-- **模型中立**：支持对接各种大模型，包括本地私有大模型（DeepSeek R1 / Llama 3 / Qwen 2 等）、国内公共大模型（通义千问 / 腾讯混元 / 字节豆包 / 百度千帆 / 智谱 AI / Kimi 等）和国外公共大模型（OpenAI / Claude / Gemini 等）；
+- **RAG 检索增强生成**：高效搭建本地 AI 知识库，支持直接上传文档 / 自动爬取在线文档，支持文本自动拆分、向量化，有效减少大模型幻觉，提升问答效果；
 - **灵活编排**：内置强大的工作流引擎、函数库和 MCP 工具调用能力，支持编排 AI 工作过程，满足复杂业务场景下的需求；
-- **无缝嵌入**：支持零编码快速嵌入到第三方业务系统，让已有系统快速拥有智能问答能力，提高用户满意度。
+- **无缝嵌入**：支持零编码快速嵌入到第三方业务系统，让已有系统快速拥有智能问答能力，提高用户满意度；
+- **模型中立**：支持对接各种大模型，包括本地私有大模型（DeepSeek R1 / Qwen 3 等）、国内公共大模型（通义千问 / 腾讯混元 / 字节豆包 / 百度千帆 / 智谱 AI / Kimi 等）和国外公共大模型（OpenAI / Claude / Gemini 等）。
 
 MaxKB 三分钟视频介绍：https://www.bilibili.com/video/BV18JypYeEkj/
 
 ## 快速开始
 
 ```
 # Linux 机器
-docker run -d --name=maxkb --restart=always -p 8080:8080 -v ~/.maxkb:/var/lib/postgresql/data -v ~/.python-packages:/opt/maxkb/app/sandbox/python-packages registry.fit2cloud.com/maxkb/maxkb
+docker run -d --name=maxkb --restart=always -p 8080:8080 -v ~/.maxkb:/var/lib/postgresql/data -v ~/.python-packages:/opt/maxkb/app/sandbox/python-packages registry.fit2cloud.com/maxkb/maxkb:v1
 
 # Windows 机器
-docker run -d --name=maxkb --restart=always -p 8080:8080 -v C:/maxkb:/var/lib/postgresql/data -v C:/python-packages:/opt/maxkb/app/sandbox/python-packages registry.fit2cloud.com/maxkb/maxkb
+docker run -d --name=maxkb --restart=always -p 8080:8080 -v C:/maxkb:/var/lib/postgresql/data -v C:/python-packages:/opt/maxkb/app/sandbox/python-packages registry.fit2cloud.com/maxkb/maxkb:v1
 
 # 用户名: admin
 # 密码: MaxKB@123..
 ```
 
 - 你也可以通过 [1Panel 应用商店](https://apps.fit2cloud.com/1panel) 快速部署 MaxKB；
 - 如果是内网环境，推荐使用 [离线安装包](https://community.fit2cloud.com/#/products/maxkb/downloads) 进行安装部署；
-- MaxKB 产品版本分为社区版和专业版，详情请参见：[MaxKB 产品版本对比](https://maxkb.cn/pricing.html)；
-- 如果您需要向团队介绍 MaxKB，可以使用这个 [官方 PPT 材料](https://maxkb.cn/download/introduce-maxkb_202503.pdf)。
+- MaxKB 不同产品产品版本的对比请参见：[MaxKB 产品版本对比](https://maxkb.cn/price)；
+- 如果您需要向团队介绍 MaxKB，可以使用这个 [官方 PPT 材料](https://fit2cloud.com/maxkb/download/introduce-maxkb_202507.pdf)。
 
 如你有更多问题，可以查看使用手册，或者通过论坛与我们交流。
 

apps/application/flow/step_node/ai_chat_step_node/impl/base_chat_node.py

@@ -40,6 +40,7 @@
 
 """
 
+
 def _write_context(node_variable: Dict, workflow_variable: Dict, node: INode, workflow, answer: str,
                    reasoning_content: str):
     chat_model = node_variable.get('chat_model')
@@ -102,7 +103,6 @@ def write_context_stream(node_variable: Dict, workflow_variable: Dict, node: INo
     _write_context(node_variable, workflow_variable, node, workflow, answer, reasoning_content)
 
 
-
 async def _yield_mcp_response(chat_model, message_list, mcp_servers):
     async with MultiServerMCPClient(json.loads(mcp_servers)) as client:
         agent = create_react_agent(chat_model, client.get_tools())
@@ -115,6 +115,7 @@ async def _yield_mcp_response(chat_model, message_list, mcp_servers):
             if isinstance(chunk[0], AIMessageChunk):
                 yield chunk[0]
 
+
 def mcp_response_generator(chat_model, message_list, mcp_servers):
     loop = asyncio.new_event_loop()
     try:
@@ -130,6 +131,7 @@ def mcp_response_generator(chat_model, message_list, mcp_servers):
     finally:
         loop.close()
 
+
 async def anext_async(agen):
     return await agen.__anext__()
 
@@ -186,7 +188,9 @@ def save_context(self, details, workflow_manage):
         self.context['answer'] = details.get('answer')
         self.context['question'] = details.get('question')
         self.context['reasoning_content'] = details.get('reasoning_content')
-        self.answer_text = details.get('answer')
+        self.context['model_setting'] = details.get('model_setting')
+        if self.node_params.get('is_result', False):
+            self.answer_text = details.get('answer')
 
     def execute(self, model_id, system, prompt, dialogue_number, history_chat_record, stream, chat_id, chat_record_id,
                 model_params_setting=None,
@@ -216,7 +220,7 @@ def execute(self, model_id, system, prompt, dialogue_number, history_chat_record
         message_list = self.generate_message_list(system, prompt, history_message)
         self.context['message_list'] = message_list
 
-        if mcp_enable and mcp_servers is not None:
+        if mcp_enable and mcp_servers is not None and '"stdio"' not in mcp_servers:
             r = mcp_response_generator(chat_model, message_list, mcp_servers)
             return NodeResult(
                 {'result': r, 'chat_model': chat_model, 'message_list': message_list,
@@ -271,6 +275,7 @@ def get_details(self, index: int, **kwargs):
             "index": index,
             'run_time': self.context.get('run_time'),
             'system': self.context.get('system'),
+            'model_setting': self.context.get('model_setting'),
             'history_message': [{'content': message.content, 'role': message.type} for message in
                                 (self.context.get('history_message') if self.context.get(
                                     'history_message') is not None else [])],