AI-Q
AI-Q 是 AI Skill Hub 本期精选Agent工作流之一。综合评分 8.0 分，整体质量较高。我们强烈推荐将其纳入你的 AI 工具库，帮助提升工作效率。
📚 深度解析
AI-Q 工作流的设计遵循"最小配置，最大复用"原则：核心逻辑已经封装好，用户只需配置自己的 API Key 和业务参数即可快速上手。工作流内置错误处理和重试机制，在网络波动或 API 限速等情况下仍能稳定运行，适合作为生产环境的自动化基础设施。
在实际部署时，建议先在测试环境中运行 3-5 次，验证各个环节的输出结果符合预期，再部署到生产环境。AI Skill Hub 评分 8.0 分，是同类 Agent 工作流中的精选推荐。
📋 工具概览
AI-Q 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
📖 中文文档
AI-Q 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排，将复杂的多步骤任务拆解为清晰的自动化流程，实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成，适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
- 可视化 Agent 工作流编排，无需编写复杂代码
- 支持多步骤自动化任务链，实现全流程无人值守
- 与外部 API、数据库和第三方服务无缝集成
- 内置错误处理与自动重试机制，保障稳定运行
- 提供可复用的自动化模板，快速在同类场景部署
- 自动化日常重复性工作，将精力集中于创造性任务
- 构建数据采集 → 处理 → 输出的完整自动化管线
- 实现跨平台、跨系统的数据流转和业务协同
# 方式一：pip 安装（推荐）
pip install aiq
# 方式二：虚拟环境安装（推荐生产环境）
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install aiq
# 方式三：从源码安装（获取最新功能）
git clone https://github.com/NVIDIA-AI-Blueprints/aiq
cd aiq
pip install -e .
# 验证安装
python -c "import aiq; print('安装成功')"
- 访问 GitHub 仓库获取工作流文件
- 在对应平台（Dify / Flowise / Make 等）中找到「导入工作流」功能
- 上传工作流文件
- 按照提示配置必要的环境变量和 API Key
- 运行测试确认流程正常后投入使用
# 命令行使用
aiq --help
# 基本用法
aiq input_file -o output_file
# Python 代码中调用
import aiq
# 示例
result = aiq.process("input")
print(result)
# aiq 配置文件示例（config.yml） app: name: "aiq" debug: false log_level: "INFO" # 运行时指定配置文件 aiq --config config.yml # 或通过环境变量配置 export AIQ_API_KEY="your-key" export AIQ_OUTPUT_DIR="./output"
简介
NVIDIA AI-Q Blueprint
🏆 BENCHMARK NOTE 🏆 To obtain results consistent with the nvidia-aiq DeepResearch Bench leaderboard and DeepResearch Bench II benchmark repository results, please use thedrb1anddrb2branches, respectively.
Overview
The NVIDIA AI-Q Blueprint is an enterprise-grade research agent built on the NVIDIA NeMo Agent Toolkit and uses LangChain Deep Agents. It gives you both quick, cited answers and in-depth, report-style research in one system, with benchmarks and evaluation harnesses so you can measure quality and improve over time.
<p align="center"> <img src="./docs/assets/AIQ-arch-light.png" alt="AI-Q Architecture" width="800"> </p>
Key features:
- Orchestration node — One node classifies intent (meta vs. research), produces meta responses (for example, greetings, capabilities), and sets research depth (shallow vs. deep).
- Shallow research — Bounded, faster researcher with tool-calling and source citation.
- Deep research — Long-running multi-step planning and research to generate a long-form citation-backed report.
- Workflow configuration — YAML configs define agents, tools, LLMs, and routing behavior so you can tune workflows without code changes.
- Modular workflows — All agents (orchestration node, shallow researcher, deep researcher, clarifier) are composable; each can run standalone or as part of the full pipeline.
- Skills and sandbox execution — Deep research can load built-in DeepAgents skills, including the
data-table-analysisworkflow, and run code-oriented work in a job-scoped Modal sandbox. - Portable agent skill — AI-Q ships
skills/aiq-research/so compatible coding harnesses can call a local AI-Q server for routed chat and async deep research jobs. - Data source registry — UI toggles and request payloads can select web, paper, enterprise, collaboration, and knowledge-layer sources per message.
- Production API and auth — REST endpoints, async job ownership, token validator entry points, and provider lifecycle hooks support authenticated deployments.
- Profiling and cost analysis — Tokenomics reports combine NAT profiler traces with pricing configuration for cost, latency, and cache analysis.
- Evaluation harnesses — Built-in benchmarks (for example, FreshQA, DeepResearch) and evaluation scripts to measure quality and iterate on prompts and agent architecture.
- Frontend options — Run through CLI, web UI, or async jobs. Refer to Getting started and Ways to run the agents.
- Deployment options - Deployment assets for a docker compose as well as helm deployment.
Prerequisites
- Python 3.11–3.13
- uv package manager
- NVIDIA API key from NVIDIA AI (for NIM models)
- Node.js 22+ and npm (optional, for web UI mode)
Optional requirements: - Tavily API key (for web search functionality) - Serper API key (for academic paper search functionality)
Note: Configure at least one data source (Tavily web search, Serper search tool, or knowledge layer) to enable research functionality.
If these optional API keys are not provided, the agent continues to operate without the corresponding search capabilities. Refer to Obtain API Keys for details.
Hardware Requirements
When using NVIDIA API Catalog (the default), inference runs on NVIDIA-hosted infrastructure and there are no local GPU requirements. The hardware references below apply only when self-hosting models via NVIDIA NIM.
| Component | Default Model | Self-Hosted Hardware Reference |
|---|---|---|
| LLM (research subagent) | nvidia/nemotron-3-nano-30b-a3b (default) or nvidia/nemotron-3-super-120b-a12b (optional) | [Nemotron 3 Nano support matrix](https://docs.nvidia.com/nim/large-language-models/latest/supported-models.html#nvidia-nemotron-3-nano), [Nemotron 3 Super support matrix](https://docs.nvidia.com/nim/large-language-models/latest/supported-models.html#nvidia-nemotron-3-super-120b-a12b) |
| LLM (intent classifier) | nvidia/nemotron-3-nano-30b-a3b | [Nemotron 3 Nano support matrix](https://docs.nvidia.com/nim/large-language-models/latest/supported-models.html#nvidia-nemotron-3-nano) |
| LLM (deep research orchestrator, planner) | openai/gpt-oss-120b | [GPT OSS support matrix](https://docs.nvidia.com/nim/large-language-models/latest/supported-models.html#gpt-oss-120b) |
| Document summary (optional) | nvidia/nemotron-mini-4b-instruct | [Nemotron Mini 4B](https://build.nvidia.com/nvidia/nemotron-mini-4b-instruct/) |
| Text embedding | nvidia/llama-nemotron-embed-vl-1b-v2 | [NeMo Retriever embedding support matrix](https://docs.nvidia.com/nim/nemo-retriever/text-embedding/latest/support-matrix.html) |
| VLM (image/chart extraction, optional) | nvidia/nemotron-nano-12b-v2-vl | [Vision language model support matrix](https://docs.nvidia.com/nim/vision-language-models/latest/support-matrix.html#nemotron-nano-12b-v2-vl) |
| Knowledge layer (Foundational RAG, optional) | -- | [RAG Blueprint support matrix](https://docs.nvidia.com/rag/latest/support-matrix.html) |
For detailed installation instructions, refer to Installation -- Hardware Requirements.
Install core with development dependencies
uv pip install -e ".[dev]"
Getting Started
Automated Setup
Run the setup script to initialize the environment:
./scripts/setup.shThis script: - Creates a Python virtual environment with uv - Installs all Python dependencies (core, frontends, benchmarks, data sources) - Installs UI dependencies (if Node.js is available)
Manual Installation
For selective installation, install packages individually:
```bash
Install frontends (pick what you need)
uv pip install -e ./frontends/cli # CLI frontend uv pip install -e ./frontends/debug # Debug console uv pip install -e ./frontends/aiq_api # Unified API (includes debug)
Install benchmarks (pick what you need)
uv pip install -e ./frontends/benchmarks/freshqa
Install data sources (pick what you need)
uv pip install -e ./sources/tavily_web_search uv pip install -e ./sources/google_scholar_paper_search uv pip install -e "./sources/knowledge_layer[llamaindex,foundational_rag]" ```
Or run directly with the NeMo Agent Toolkit CLI (dotenv loads deploy/.env into the environment)
dotenv -f deploy/.env run nat run --config_file configs/config_cli_default.yml --input "How do I install CUDA?" ```
The CLI frontend source is in frontends/cli/.
No-auth local setup (LlamaIndex default)
docker compose --env-file ../.env -f docker-compose.yaml up -d --build
To select a different backend config, set BACKEND_CONFIG in deploy/.env, for example:
Step 1: Install the dataset
The dataset files are not included in the repository. We have included a script to retrieve them from the Deep Research Bench Github Repository and format them for the NeMo Agent Toolkit evaluator.
To download the dataset files, run the following script:
python frontends/benchmarks/deepresearch_bench/scripts/download_drb_dataset.pyCreate and activate virtual environment
uv venv --python 3.13 .venv source .venv/bin/activate
Set Up Environment Variables
Create a .env file in deploy/ directory:
cp deploy/.env.example deploy/.envReplace your API keys.
Note: Depending on your usecase, deep research report quality can be enhanced by enabling searching across academic research papers. We use Serper for this. If you want to use paper search, follow the steps in the Customization guide to enable it.
Configuration Files
The configs/ directory holds YAML workflow configs that define agents, tools, LLMs, and routing. Use the one that matches your run mode and data sources:
| Config | Models | Description |
|---|---|---|
config_cli_default.yml | Nemotron 3 Nano 30B, GPT-OSS 120B | CLI default. Web search; optional paper search (requires SERPER_API_KEY); no knowledge retrieval. Nemotron Super is commented out but can be enabled for higher quality. |
config_web_default_llamaindex.yml | Nemotron 3 Nano 30B, GPT-OSS 120B, Nemotron Mini 4B | Web default. LlamaIndex knowledge retrieval; web search; optional paper search (requires SERPER_API_KEY). Nemotron Super is commented out but can be enabled for higher quality. |
config_web_frag.yml | Nemotron 3 Nano 30B, GPT-OSS 120B | Web + Foundational RAG (external RAG server). Helm default. See [RAG Blueprint](https://github.com/NVIDIA-AI-Blueprints/rag/tree/main) for an example RAG deployment. Nemotron Super is commented out but can be enabled for higher quality. |
config_frontier_models.yml | GPT-5.2 (orchestrator/planner), Nemotron 3 Nano 30B, Nemotron Mini 4B | Hybrid: frontier orchestrator/planner, open researcher. LlamaIndex; web search; optional paper search (requires SERPER_API_KEY). Requires OPENAI_API_KEY. Nemotron Super is commented out but can be enabled for higher quality. |
Activate the virtual environment
source .venv/bin/activate
BACKEND_CONFIG=/app/configs/config_web_frag.yml
```
For more details, refer to: - deploy/compose/README.md
Optional: Phoenix Tracing
If your config enables Phoenix tracing, start the Phoenix server before running nat eval.
Start server (separate terminal):
source .venv/bin/activate
phoenix serveFor detailed benchmark documentation, refer to: - Deep Research Bench README - FreshQA README
Obtain API Keys
| API | Environment Variable | Purpose | Required |
|---|---|---|---|
| NVIDIA API | NVIDIA_API_KEY | LLM inference through NIM | Yes |
| Tavily | TAVILY_API_KEY | Web search | No (if not specified, agent continues without web search) |
| Serper | SERPER_API_KEY | Academic paper search | No (if not specified, agent continues without paper search) |
Obtain an NVIDIA API Key
- Sign in to NVIDIA Build
- Click on any model, then select "Deploy" > "Get API Key" > "Generate Key"
Obtain a Tavily API Key
- Sign in to Tavily
- Navigate to your dashboard
- Generate an API key
Obtain a Serper API Key
- Sign in to Serper
- Generate an API key from your dashboard
Command-line interface (CLI)
The CLI provides an interactive research assistant in your terminal:
```bash
Software Components
The following are used by this project in the default configuration:
- NVIDIA NeMo Agent Toolkit
- NVIDIA nemotron-3-nano-30b-a3b (agents, researcher)
- NVIDIA nemotron-3-super-120b-a12b (optional, compatible but Build API has limited availability due to high demand)
- NVIDIA nemotron-3-nano-30b-a3b (intent classifier)
- GPT-OSS-120B (agents)
- NVIDIA nemotron-mini-4b-instruct (document summary, if used)
- NVIDIA llama-nemotron-embed-vl-1b-v2 (embedding model for llamaindex knowledge layer implementation, if used)
- NVIDIA nemotron-nano-12b-v2-vl (vision-language model for llamaindex knowledge layer implementation, if used)
- Tavily Search API for web search
- Serper Search API for paper search (Google Scholar)
Evaluating the Workflow
The frontends/benchmarks/ directory contains evaluation pipelines for assessing agent performance.
高质量的开源AI工作流示例
- 需要 aiq 解决具体问题的开发者与运营人员
- 先在测试环境跑通最小用例，再接入生产数据
- API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
- Python 依赖冲突：建议用 venv / uv 隔离环境
- 云端托管：可放在 Vercel / Railway / Fly.io 等 PaaS 平台
⚡ 核心功能
- 可视化 Agent 工作流编排，无需编写复杂代码
- 支持多步骤自动化任务链，实现全流程无人值守
- 与外部 API、数据库和第三方服务无缝集成
- 内置错误处理与自动重试机制，保障稳定运行
- 提供可复用的自动化模板，快速在同类场景部署
- 需要 aiq 解决具体问题的开发者与运营人员
- 先在测试环境跑通最小用例，再接入生产数据
- API key 直接提交到 git 仓库（请用 .env 并加入 .gitignore）
- Python 依赖冲突：建议用 venv / uv 隔离环境
👥 适合人群
🎯 使用场景
- 自动化日常重复性工作，将精力集中于创造性任务
- 构建数据采集 → 处理 → 输出的完整自动化管线
- 实现跨平台、跨系统的数据流转和业务协同
⚖️ 优点与不足
- +Apache-2.0 协议，可免费商用
- +大幅减少重复性人工操作
- +可视化流程，清晰直观
- +可扩展性强，支持复杂场景
- −初始配置和调试需投入一定时间
- −强依赖外部服务的稳定性
- −复杂场景需具备一定技术基础
AI Skill Hub 为第三方内容聚合平台，本页面信息基于公开数据整理，不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后，再部署至生产环境，并做好必要的安全评估。
✅ Apache 2.0 — 宽松开源协议，可商用，需保留版权声明和 NOTICE 文件，含专利授权条款。
🔗 相关工具推荐
❓ 常见问题 FAQ
经综合评估，AI-Q 在Agent工作流赛道中表现稳健，质量优秀。如果你已有明确的使用需求，可以直接上手体验；如果还在评估阶段，建议对比同类工具后再做决策。
| 原始名称 | aiq |
| 原始描述 | 开源AI工作流：The AI-Q NVIDIA Blueprint is an open reference example for building intelligent 。⭐735 · Python |
| Topics | AINVIDIAPython |
| GitHub | https://github.com/NVIDIA-AI-Blueprints/aiq |
| License | Apache-2.0 |
| 语言 | Python |
收录时间：2026-06-15 · 更新时间：2026-06-16 · License：Apache-2.0 · AI Skill Hub 不对第三方内容的准确性作法律背书。