{"slug": "rig-agentic-workflows-in-rust", "title": "Rig – Agentic Workflows in Rust", "summary": "Rig, a Rust library for building LLM-powered applications, has been released with support for agentic workflows, multi-turn streaming, and 20+ model providers under a unified interface. The library, developed by 0xPlaygrounds, includes a classic agent runtime, 10+ vector store integrations, and browser-WASM support, with users including St. Jude, Coral Protocol, and Nethermind.", "body_md": "[📑 Docs](https://rig.rs/docs)\n•\n[🌐 Website](https://rig.rs)\n•\n[🤝 Contribute](https://github.com/0xPlaygrounds/rig/issues/new)\n•\n[✍🏽 Blogs](https://rig.rs/docs/guides)\n•\n\n✨ If you would like to help spread the word about Rig, please consider starring the repo!\n\nWarning\n\nHere be dragons! As we plan to ship a torrent of features in the following months, future updates **will** contain **breaking changes**. With Rig evolving, we'll annotate changes and highlight migration paths as we encounter them.\n\nRig is a Rust library for building scalable, modular, and ergonomic **LLM-powered** applications.\n\nMore information about this crate can be found in the [official](https://rig.rs/docs) and [crate](https://docs.rs/rig/latest/rig/) API reference documentation.\n\n- Agentic workflows that can handle multi-turn streaming and prompting\n- A classic agent runtime enabled by default\n- Full\n[GenAI Semantic Convention](https://opentelemetry.io/docs/specs/semconv/gen-ai/)compatibility - 20+ model providers, all under one singular unified interface\n- 10+ vector store integrations, all under one singular unified interface\n- Full support for LLM completion and embedding workflows\n- Support for transcription, audio generation and image generation model capabilities\n- Integrate LLMs in your app with minimal boilerplate\n- Browser-WASM (\n`wasm32-unknown-unknown`\n\n) support for the portable core and classic runtime — see[target support](/0xPlaygrounds/rig/blob/main/crates/rig-agent/README.md#target-support)for the full matrix (WASI is not supported;`rig-rmcp`\n\n/MCP is native-only)\n\nRig separates portable provider/backend contracts from agent orchestration:\n\n`rig-core`\n\ncontains provider-neutral messages, completion models, portable tools, memory and vector-store contracts, and built-in provider mappings.`rig-agent`\n\ncontains the classic builder, prompt/streaming traits, typed hooks, contextual tools, extraction, and the serializable`AgentRun`\n\nstate machine. It remains enabled by default.\n\nThe root `rig`\n\nfacade re-exports both at their familiar paths, so most code\ndepends only on `rig`\n\n.\n\nBelow is a non-exhaustive list of companies and people who are using Rig:\n\n[St Jude](https://www.stjude.org/)- Using Rig for a chatbot utility as part of, a genomics visualisation tool.`proteinpaint`\n\n[Coral Protocol](https://www.coralprotocol.org/)- Using Rig extensively, both internally as well as part of the[Coral Rust SDK.](https://github.com/Coral-Protocol/coral-rs)[VT Code](https://github.com/vinhnx/vtcode)- VT Code is a Rust-based terminal coding agent with semantic code intelligence via Tree-sitter and ast-grep. VT Code uses`rig`\n\nfor simplifying LLM calls and implementing the model picker.[Con](https://github.com/nowledge-co/con)- Con is a GPU-accelerated terminal emulator with a built-in AI agent harness. It uses Rig as the provider abstraction layer for its integrated coding agents.[Dria](https://dria.co/)- a decentralised AI network. Currently using Rig as part of their[compute node.](https://github.com/firstbatchxyz/dkn-compute-node)[Nethermind](https://www.nethermind.io/)- Using Rig as part of their[Neural Interconnected Nodes Engine](https://github.com/NethermindEth/nine)framework.[Neon](https://neon.com)- Using Rig for their[app.build](https://github.com/neondatabase/appdotbuild-agent)V2 reboot in Rust.[Listen](https://github.com/piotrostr/listen)- A framework aiming to become the go-to framework for AI portfolio management agents. Powers[the Listen app.](https://app.listen-rs.com/)[Cairnify](https://cairnify.com/)- helps users find documents, links, and information instantly through an intelligent search bar. Rig provides the agentic foundation behind Cairnify’s AI search experience, enabling tool-calling, reasoning, and retrieval workflows.[Ryzome](https://ryzome.ai)- Ryzome is a visual AI workspace that lets you build interconnected canvases of thoughts, research, and AI agents to orchestrate complex knowledge work.[deepwiki-rs](https://github.com/sopaco/deepwiki-rs)- Turn code into clarity. Generate accurate technical docs and AI-ready context in minutes—perfectly structured for human teams and intelligent agents.[Cortex Memory](https://github.com/sopaco/cortex-mem)- The production-ready memory system for intelligent agents. A complete solution for memory management, from extraction and vector search to automated optimization, with a REST API, MCP, CLI, and insights dashboard out-of-the-box.[Ironclaw](https://github.com/nearai/ironclaw)- A secure personal AI assistant[ilert](https://www.ilert.com/)- Incident management & alerting platform. Uses Rig as the multi-provider abstraction in its agentic LLM proxy powering ilert AI.[Archestra](https://github.com/archestra-ai/archestra)- MCP-native secure AI platform. Uses Rig in its agentic benchmark.\n\nFor a curated list of Rig projects, libraries, tools, articles, and production users, check out [awesome-rig](https://github.com/0xPlaygrounds/awesome-rig).\n\nAre you also using Rig? [Open an issue](https://www.github.com/0xPlaygrounds/rig/issues) to have your name added!\n\nUse the root `rig`\n\nfacade when you want feature-gated access to companion crates,\nor use `rig-core`\n\ndirectly when you only need the core provider abstractions.\n\n```\ncargo add rig\n# or: cargo add rig-core\nuse rig::prelude::*;\nuse rig::providers::openai;\n\n#[tokio::main]\nasync fn main() -> Result<(), anyhow::Error> {\n    // Create OpenAI client\n    let client = openai::Client::from_env()?;\n\n    // Create agent with a single context prompt\n    let comedian_agent = client\n        .agent(openai::GPT_5_2)\n        .preamble(\"You are a comedian here to entertain the user using humour and jokes.\")\n        .build();\n\n    // Prompt the agent and print the response\n    let response = comedian_agent.prompt(\"Entertain me!\").await?;\n\n    println!(\"{response}\");\n\n    Ok(())\n}\n```\n\nNote using `#[tokio::main]`\n\nrequires you enable tokio's `macros`\n\nand `rt-multi-thread`\n\nfeatures\nor just `full`\n\nto enable all features (`cargo add tokio --features macros,rt-multi-thread`\n\n).\n\nYou can find more examples in each crate's `examples`\n\ndirectory (for example, [ examples](/0xPlaygrounds/rig/blob/main/examples)). Provider-specific integration coverage lives under\n\n[, with cassette-backed tests that replay offline by default and live-only tests kept separate when real provider APIs are still required. See](/0xPlaygrounds/rig/blob/main/tests/providers)\n\n`tests/providers`\n\n[for test target, replay, record, and cassette safety commands. More detailed use case walkthroughs are regularly published on our](/0xPlaygrounds/rig/blob/main/tests/README.md)\n\n`tests/README.md`\n\n[Dev.to Blog](https://dev.to/0thtachi)and added to Rig's official documentation at\n\n[rig.rs/docs](https://rig.rs/docs).\n\nThe root `rig`\n\nfacade exposes companion crates behind one feature per integration:\n\n```\nrig = { version = \"0.36.0\", features = [\"lancedb\", \"fastembed\"] }\n```\n\n| Integration | Crate | Feature | Module path |\n|---|---|---|---|\n| AWS Bedrock |\n`rig-bedrock` |\n\n`bedrock`\n\n`rig::bedrock`\n\n`rig-s3vectors`\n\n`s3vectors`\n\n`rig::s3vectors`\n\n`rig-candle`\n\n`candle`\n\n`rig::candle`\n\n`rig-vectorize`\n\n`vectorize`\n\n`rig::vectorize`\n\n`rig-fastembed`\n\n`fastembed`\n\n`rig::fastembed`\n\n`rig-gemini-grpc`\n\n`gemini-grpc`\n\n`rig::gemini_grpc`\n\n`rig-vertexai`\n\n`vertexai`\n\n`rig::vertexai`\n\n`rig-helixdb`\n\n`helixdb`\n\n`rig::helixdb`\n\n`rig-lancedb`\n\n`lancedb`\n\n`rig::lancedb`\n\n`rig-memory`\n\n`memory`\n\n`rig::memory`\n\n`rig-milvus`\n\n`milvus`\n\n`rig::milvus`\n\n`rig-mongodb`\n\n`mongodb`\n\n`rig::mongodb`\n\n`rig-neo4j`\n\n`neo4j`\n\n`rig::neo4j`\n\n`rig-postgres`\n\n`postgres`\n\n`rig::postgres`\n\n`rig-qdrant`\n\n`qdrant`\n\n`rig::qdrant`\n\n`rig-scylladb`\n\n`scylladb`\n\n`rig::scylladb`\n\n`rig-sqlite`\n\n`sqlite`\n\n`rig::sqlite`\n\n`rig-surrealdb`\n\n`surrealdb`\n\n`rig::surrealdb`\n\n`rig::memory`\n\nis available without the `memory`\n\nfeature; it contains the core\nconversation memory traits and in-memory backend re-exported from `rig-core`\n\n.\nEnabling `features = [\"memory\"]`\n\nadds reusable history-shaping policy types from\nthe `rig-memory`\n\ncompanion crate to the same module.\n\nWe also have some other associated crates that have additional functionality you may find helpful when using Rig:\n\n`rig-onchain-kit`\n\n- the[Rig Onchain Kit.](https://github.com/0xPlaygrounds/rig-onchain-kit)Intended to make interactions between Solana/EVM and Rig much easier to implement.", "url": "https://wpnews.pro/news/rig-agentic-workflows-in-rust", "canonical_source": "https://github.com/0xplaygrounds/rig", "published_at": "2026-08-29 13:22:03+00:00", "updated_at": "2026-08-29 13:48:38.879034+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "developer-tools", "ai-agents"], "entities": ["Rig", "0xPlaygrounds", "St. Jude", "Coral Protocol", "Nethermind", "Neon", "Dria", "Cairnify"], "alternates": {"html": "https://wpnews.pro/news/rig-agentic-workflows-in-rust", "markdown": "https://wpnews.pro/news/rig-agentic-workflows-in-rust.md", "text": "https://wpnews.pro/news/rig-agentic-workflows-in-rust.txt", "jsonld": "https://wpnews.pro/news/rig-agentic-workflows-in-rust.jsonld"}}