futureagi-sdk
starsforks
How to install futureagi-sdk?
future-agi/futureagi-sdk
$git clone https://github.com/future-agi/futureagi-sdkThe open-source SDK for AI evaluation, observability, and optimization
README.md
| 1 | <div align="center"> |
| 2 | |
| 3 |  |
| 4 | |
| 5 | # Future AGI SDK |
| 6 | |
| 7 | **The open-source SDK for AI evaluation, observability, and optimization** |
| 8 | |
| 9 | [](https://pypi.org/project/futureagi/) |
| 10 | [](https://pepy.tech/project/futureagi) |
| 11 | [](https://pypi.org/project/futureagi/) |
| 12 | [](LICENSE) |
| 13 | [](https://github.com/future-agi/futureagi-sdk) |
| 14 | |
| 15 | [📖 Docs](https://docs.futureagi.com) • [🌐 Website](https://www.futureagi.com) • [💬 Community](https://www.linkedin.com/company/futureagi) • [🎯 Dashboard](https://app.futureagi.com) |
| 16 | |
| 17 | </div> |
| 18 | |
| 19 | --- |
| 20 | |
| 21 | ## 📖 Table of Contents |
| 22 | |
| 23 | - [What is Future AGI?](#-what-is-future-agi) |
| 24 | - [Installation](#-installation) |
| 25 | - [Authentication](#-authentication) |
| 26 | - [30-Second Examples](#-30-second-examples) |
| 27 | - [Quick Start](#-quick-start) |
| 28 | - [Dataset Management](#-dataset-management) |
| 29 | - [Prompt Workbench](#-prompt-workbench) |
| 30 | - [Knowledge Base (RAG)](#-knowledge-base-rag) |
| 31 | - [How It Works](#️-how-it-works) |
| 32 | - [Core Use Cases](#-core-use-cases) |
| 33 | - [Real-World Use Cases](#-real-world-use-cases) |
| 34 | - [Why Choose Future AGI?](#-why-choose-future-agi) |
| 35 | - [Supported Integrations](#-supported-integrations) |
| 36 | - [Documentation](#-documentation) |
| 37 | - [Language Support](#-language-support) |
| 38 | - [Support & Community](#-support--community) |
| 39 | - [Contributing](#-contributing) |
| 40 | - [Testimonials](#-testimonials) |
| 41 | - [Roadmap](#-roadmap) |
| 42 | - [Troubleshooting & FAQ](#-troubleshooting--faq) |
| 43 | |
| 44 | --- |
| 45 | |
| 46 | ## 🚀 What is Future AGI? |
| 47 | |
| 48 | Your agent passed every eval. Then it hallucinated a refund policy that doesn't exist. Future AGI gives you the tools to catch that — datasets, prompt versioning, knowledge bases, evaluations, and guardrails. One SDK, one feedback loop. |
| 49 | |
| 50 | ```bash |
| 51 | # Get started in 30 seconds |
| 52 | pip install futureagi |
| 53 | export FI_API_KEY="your_key" |
| 54 | export FI_SECRET_KEY="your_secret" |
| 55 | ``` |
| 56 | |
| 57 | **👉 [Get Free API Keys](https://app.futureagi.com/signup) • [View Live Demo](https://docs.futureagi.com) • [Read Quick Start Guide](https://docs.futureagi.com/get-started/quickstart)** |
| 58 | |
| 59 | ### ✨ Key Features |
| 60 | |
| 61 | - **🎯 Evaluations** — 50+ metrics, LLM-as-judge, and custom rubrics powered by the Critique AI agent |
| 62 | - **⚡ Guardrails** — Real-time safety checks with sub-100ms latency |
| 63 | - **📊 Datasets** — Programmatically create, version, and manage training and test datasets |
| 64 | - **🎨 Prompt Workbench** — Version control, A/B testing, and deployment labels for prompts |
| 65 | - **📚 Knowledge Base** — Document management and retrieval for RAG applications |
| 66 | - **📈 Analytics** — Model performance, token costs, and behavior insights |
| 67 | - **🤖 Simulate** — Test your AI system against realistic scenarios before users hit it |
| 68 | - **🔍 Observability** — OpenTelemetry-native tracing across 50+ frameworks |
| 69 | |
| 70 | |
| 71 | --- |
| 72 | |
| 73 | ## 📦 Installation |
| 74 | |
| 75 | ### Python |
| 76 | ```bash |
| 77 | pip install futureagi |
| 78 | ``` |
| 79 | |
| 80 | ### TypeScript/JavaScript |
| 81 | ```bash |
| 82 | npm install @future-agi/sdk |
| 83 | # or |
| 84 | pnpm add @future-agi/sdk |
| 85 | ``` |
| 86 | |
| 87 | **Requirements:** Python >= 3.10 | Node.js >= 14 |
| 88 | |
| 89 | --- |
| 90 | |
| 91 | ## 🔑 Authentication |
| 92 | |
| 93 | Get your API credentials from the [Future AGI Dashboard](https://app.futureagi.com): |
| 94 | |
| 95 | ```bash |
| 96 | export FI_API_KEY="your_api_key" |
| 97 | export FI_SECRET_KEY="your_secret_key" |
| 98 | ``` |
| 99 | |
| 100 | Or set them programmatically: |
| 101 | |
| 102 | ```python |
| 103 | import os |
| 104 | os.environ["FI_API_KEY"] = "your_api_key" |
| 105 | os.environ["FI_SECRET_KEY"] = "your_secret_key" |
| 106 | ``` |
| 107 | |
| 108 | ## 🎯 Quick Start |
| 109 | |
| 110 | ### 📊 Dataset Management |
| 111 | |
| 112 | Create and manage datasets with built-in evaluations: |
| 113 | |
| 114 | ```python |
| 115 | from fi.datasets import Dataset |
| 116 | from fi.datasets.types import ( |
| 117 | Cell, Column, DatasetConfig, DataTypeChoices, |
| 118 | ModelTypes, Row, SourceChoices |
| 119 | ) |
| 120 | |
| 121 | # Create a new dataset |
| 122 | config = DatasetConfig(name="qa_dataset", |
People who liked this also viewed
6 picksType
- promptpexTest Generation for PromptsPromptsMay 2026
- langfuse🪢 Open source AI engineering platform: LLM evals, observability, metrics, prompt management, playground, datasets. Integrates with OpenTelemetry, LangChain, OpenAI SDK, LiteLLM, and more. 🍊YC W23 PromptsJul 2026
- promptflowBuild high-quality LLM apps - from prototyping, testing to production deployment and monitoring.PromptsJul 2026
- pezzo🕹️ Open-source, developer-first LLMOps platform designed to streamline prompt design, version management, instant delivery, collaboration, troubleshooting, observability and more.PromptsMar 2026
- ChainForgeAn open-source visual programming environment for battle-testing prompts to LLMs.PromptsJun 2026
- PromptFuzzPromtFuzz is an automated tool that generates high-quality fuzz drivers for libraries via a fuzz loop constructed on mutating LLMs' prompts.PromptsMay 2026
