llm.rb 11.3.0 → 12.0.0

data/README.md

@@ -1,135 +1,86 @@
 <p align="center">
-  <a href="https://github.com/llmrb/llm.rb">
-    <img src="https://github.com/llmrb/llm.rb/raw/main/llm.png" width="200" height="200" border="0" alt="llm.rb">
+  <a href="https://r.uby.dev">
+    <img
+      src="https://github.com/r-uby-dev/llm.rb/raw/main/rubydev.svg"
+      width="400"
+      height="200"
+      border="0"
+      alt="a r.uby.dev project"
+     >
   </a>
 </p>
-<p align="center">
-  <a href="https://llmrb.github.io/llm.rb">
-    <img src="https://img.shields.io/badge/docs-llmrb.github.io-blue.svg" alt="Official llm.rb website">
-  </a>
-  <a href="https://opensource.org/license/0bsd">
-    <img src="https://img.shields.io/badge/License-0BSD-orange.svg?" alt="License">
-  </a>
-  <a href="https://github.com/llmrb/llm.rb/tags">
-    <img src="https://img.shields.io/badge/version-11.3.0-green.svg?" alt="Version">
-  </a>
-</p>
-
-## About
-
-llm.rb is Ruby's most capable AI runtime.
-
-It runs on Ruby's standard library by default. loads optional pieces
-only when needed, and offers a single runtime for providers, agents,
-tools, skills, MCP, A2A (Agent2Agent), RAG (vector stores & embeddings),
-streaming, files, and persisted state.
 
-It supports OpenAI, OpenAI-compatible endpoints, Anthropic, Google
-Gemini, DeepSeek, xAI, Z.ai, AWS Bedrock, Ollama, and llama.cpp. It
-also includes built-in ActiveRecord and Sequel support, plus concurrent
-tool execution through threads, tasks (via async gem), fibers, ractors,
-and fork (via xchan.rb gem).
+> A [r.uby.dev](https://r.uby.dev) project.
 
-## Services
+Welcome to the canonical llm.rb repository.
 
-The llm.rb runtime and its forks
-([mruby-llm](https://github.com/llmrb/mruby-llm),
-[wasm-llm](https://github.com/llmrb/wasm-llm))
-power a growing family of AI applications, and
-services. The following applications are publicly
-accessible over SSH and are free to try. No account
-required. Nothing to install.
+llm.rb is not a library, framework or toolkit but an advanced runtime
+for building highly capable AI applications on CRuby. By default
+it has zero runtime dependencies although certain functionality &ndash;
+such as ActiveRecord support &ndash; require optional dependencies
+that are opt-in.
 
-#### matz - the mruby expert
+## Features
 
-> ssh matz@r.uby.dev
+The runtime supports OpenAI, OpenAI-compatible endpoints, Anthropic, Google
+Gemini, DeepSeek, DeepInfra, xAI, Z.ai, AWS Bedrock, Ollama, and llama.cpp.
+It has first-class support for streaming, tool calls,  MCP
+and A2A, embeddings, vector stores and the RAG pattern.
 
-See [https://r.uby.dev/matz](https://r.uby.dev/matz) for more information.
+There are multiple HTTP backends to choose from, tools can be run concurrently
+or in parallel via threads, async tasks, fibers, ractors, and fork, and it is
+also possible to make a tool call while the model is still streaming.
 
-#### robert - the freebsd expert
+The runtime builds on top of three core concepts: providers, contexts, and agents,
+so once you learn the fundamentals, everything else falls into place naturally. And once
+you learn llm.rb, you will also be able to use <a href="https://r.uby.dev/mruby-llm">mruby-llm</a> and
+<a href="https://r.uby.dev/wasm-llm">wasm-llm</a> because the API is pretty much identical.
 
-> ssh robert@4.4bsd.dev
+## Install
 
-See [https://4.4bsd.dev/robert](https://4.4bsd.dev/robert) for more information.
+```bash
+gem install llm.rb
+```
 
 ## Quick start
 
-#### LLM::Context
-
-The
-[LLM::Context](https://0x1eef.github.io/x/llm.rb/LLM/Context.html)
-object is at the heart of the runtime. Almost all other features build
-on top of it. It is a low-level interface to a model, and requires tool
-execution to be managed manually. The
-[LLM::Agent](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html)
-class is almost the same as
-[LLM::Context](https://0x1eef.github.io/x/llm.rb/LLM/Context.html)
-but it manages tool execution for you - we'll cover agents next:
-
-```ruby
-require "llm"
-
-llm = LLM.openai(key: ENV["KEY"])
-ctx = LLM::Context.new(llm, stream: $stdout)
-ctx.talk "Hello world"
-```
-
 #### LLM::Agent
 
-The
-[LLM::Agent](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html)
-object is implemented on top of
-[LLM::Context](https://0x1eef.github.io/x/llm.rb/LLM/Context.html).
-It provides the same interface, but manages tool execution for you. It
-also has builtin features such as a loop guard that detects repeated
-tool call patterns, and another guard that detects infinite tool call
-loops. Both guards advise the model to change course rather than raise
-an error:
+The [`LLM::Agent`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html) class is the default high-level interface,
+and it is recommended for most use-cases. It manages tool execution
+automatically, guards against infinite loops, manages conversation
+state, and much more.
 
 ```ruby
 require "llm"
 
-llm = LLM.openai(key: ENV["KEY"])
+llm = LLM.deepseek(key: ENV["KEY"])
 agent = LLM::Agent.new(llm, stream: $stdout)
 agent.talk "Hello world"
 ```
 
-#### Agents (Advanced)
+#### LLM::Context
 
-An agent can be configured to require confirmation before a tool is
-executed. When a matching tool is called, llm.rb runs
-`on_tool_confirmation`. That callback must decide whether to cancel the
-tool call or approve it and execute it by calling
-`fn.spawn(strategy).wait`, and it must always return an instance of
-[`LLM::Function::Return`](https://0x1eef.github.io/x/llm.rb/LLM/Function/Return.html):
+The [`LLM::Context`](https://r.uby.dev/api-docs/llm.rb/LLM/Context.html) class is at the heart of the runtime
+and it is what [`LLM::Agent`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html) uses under the hood.
+It requires that the tool call loop be managed manually -
+sometimes that can be useful, but usually for advanced use-cases.
+If you're new to llm.rb, try [`LLM::Agent`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html) first.
 
 ```ruby
 require "llm"
 
-class Agent < LLM::Agent
-  tools DeleteFile
-  confirm "delete-file"
-
-  def on_tool_confirmation(fn, strategy)
-    path = fn.arguments.path
-    if path.start_with?("/tmp/")
-      fn.spawn(strategy).wait
-    else
-      fn.cancel(reason: "Deletion requires approval")
-    end
-  end
-end
-
-llm = LLM.openai(key: ENV["KEY"])
-Agent.new(llm, stream: $stdout).talk("Delete /tmp/example.txt.")
+llm = LLM.deepseek(key: ENV["KEY"])
+ctx = LLM::Context.new(llm, stream: $stdout)
+ctx.talk "Hello world"
 ```
 
-#### Tools
+#### LLM::Tool
 
-The
-[LLM::Tool](https://0x1eef.github.io/x/llm.rb/LLM/Tool.html)
-class can be subclassed to implement your own tools that can extend the
-abilities of a model:
+Subclasses of [`LLM::Tool`](https://r.uby.dev/api-docs/llm.rb/LLM/Tool.html) are plain Ruby classes with
+an optional set of typed parameters. <br> The model can choose to
+call them on your behalf, and they're one of the most powerful features
+for extending the feature set or abilities of a model.
 
 ```ruby
 class ReadFile < LLM::Tool
@@ -144,632 +95,243 @@ class ReadFile < LLM::Tool
 end
 ```
 
-#### MCP
-
-The
-[LLM::MCP](https://0x1eef.github.io/x/llm.rb/LLM/MCP.html)
-object lets llm.rb use tools provided by an MCP server. Those tools are
-exposed through the same runtime as local tools, so you can pass them
-to either
-[LLM::Context](https://0x1eef.github.io/x/llm.rb/LLM/Context.html)
-or
-[LLM::Agent](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html).
-In this example, the MCP server runs over stdio and
-[LLM::Agent](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html)
-manages the tool loop. For **stdio**, `mcp.session` is the preferred
-pattern because it keeps one MCP session alive across discovery and
-tool calls:
-
-```ruby
-require "llm"
-
-llm = LLM.openai(key: ENV["KEY"])
-mcp = LLM::MCP.stdio(argv: ["ruby", "server.rb"])
-
-mcp.session do
-  agent = LLM::Agent.new(llm, stream: $stdout, tools: mcp.tools)
-  agent.talk "Use the available tools to inspect the environment."
-end
-```
-
-MCP can also be used without `session`. Although it works it is generally
-not recommended for the **stdio** transport because it is inefficient
-to start and stop a fresh MCP process for tool discovery and each tool
-call:
-
-```ruby
-require "llm"
-
-llm = LLM.openai(key: ENV["KEY"])
-mcp = LLM::MCP.stdio(argv: ["ruby", "server.rb"])
-
-agent = LLM::Agent.new(llm, tools: mcp.tools)
-agent.talk("Use the available tools to inspect the environment.")
-```
-
-The HTTP transport can be used with or without the `session` method,
-and unlike the stdio transport it can remain efficient without the
-`session` method through a persistent connection pool that is available
-through the
-[LLM::Transport.net_http_persistent](https://0x1eef.github.io/x/llm.rb/LLM/Transport.html#method-c-net_http_persistent)
-transport:
-
-```ruby
-require "llm"
-
-llm = LLM.openai(key: ENV["KEY"])
-mcp = LLM::MCP.http(
-  url: "https://remote-mcp.example.com",
-  transport: :net_http_persistent
-)
-
-agent = LLM::Agent.new(llm, tools: mcp.tools)
-agent.talk("Use the available tools to inspect the environment.")
-```
-
-#### A2A (Agent 2 Agent)
-
-The
-[LLM::A2A](https://0x1eef.github.io/x/llm.rb/LLM/A2A.html)
-object lets llm.rb use skills provided by a remote A2A agent. Those
-skills are exposed through the same runtime as local tools, so you can
-pass them to either
-[LLM::Context](https://0x1eef.github.io/x/llm.rb/LLM/Context.html)
-or
-[LLM::Agent](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html).
-
-Use remote skills as local tools:
-
-```ruby
-require "llm"
-
-a2a = LLM::A2A.rest(
-  url: "https://remote-agent.example.com",
-  headers: {"Authorization" => "Bearer token"}
-)
-llm = LLM.openai(key: ENV["KEY"])
-agent = LLM::Agent.new(llm, tools: a2a.skills)
-agent.talk "Analyze this CSV and summarize the trends."
-```
-
-Use persistent HTTP connections:
-
-```ruby
-require "llm"
-
-a2a = LLM::A2A.rest(
-  url: "https://remote-agent.example.com",
-  transport: :net_http_persistent
-)
-```
-
-For more on direct messaging, task operations, push notification
-configs, and JSON-RPC, see the
-[LLM::A2A API docs](https://0x1eef.github.io/x/llm.rb/LLM/A2A.html).
-
-#### Transports
-
-Providers use Ruby's standard library Net::HTTP transport by default.
-You can opt into persistent Net::HTTP connections with `persistent: true`,
-or provide a transport shortcut when you want a different backend.
-`transport: :curb` uses libcurl through the optional `curb` gem.
-
-Custom transports can implement the
-[LLM::Transport](https://0x1eef.github.io/x/llm.rb/LLM/Transport.html)
-interface and receive transport-agnostic
-[LLM::Transport::Request](https://0x1eef.github.io/x/llm.rb/LLM/Transport/Request.html)
-objects from providers.
-
-```ruby
-require "llm"
-
-llm = LLM.openai(key: ENV["KEY"], persistent: true)
-llm = LLM.openai(key: ENV["KEY"], transport: :net_http_persistent)
-llm = LLM.openai(key: ENV["KEY"], transport: :curb)
-```
-
-#### Skills
-
-Skills are reusable instructions loaded from a `SKILL.md` directory. They let
-you package behavior and tool access together, and they plug into the
-same runtime as tools, agents, MCP, and A2A. When a skill runs, llm.rb
-spawns a subagent with the skill instructions, access to only the tools
-listed in the skill, and recent conversation context:
-
-```yaml
----
-name: release
-description: Prepare a release
-tools: ["search-docs", "git"]
----
-
-## Task
-
-Review the release state, summarize what changed, and prepare the release.
-```
-
-```ruby
-require "llm"
-
-class ReleaseAgent < LLM::Agent
-  model "gpt-5.4-mini"
-  skills "./skills/release"
-end
-
-llm = LLM.openai(key: ENV["KEY"])
-ReleaseAgent.new(llm, stream: $stdout).talk("Prepare the next release.")
-```
-
-A skill can also have its sub-agent inherit the parents tools through the
-`inherit` directive. The `inherit` directive has coverage for the "classic"
-tools (a subclass of [LLM::Tool](https://0x1eef.github.io/x/llm.rb/LLM/Tool.html)),
-MCP tools, and A2A tools that a parent context or agent has access to:
-
-```yaml
----
-  name: release
-  description: Prepare a release
-  tools: inherit
----
-```
-
 #### LLM::Stream
 
-The
-[LLM::Stream](https://0x1eef.github.io/x/llm.rb/LLM/Stream.html)
-object lets you observe output and runtime events as they happen. You
-can subclass it to handle streamed content in your own application:
+Streams can be simple IO objects or subclasses of
+[`LLM::Stream`](https://r.uby.dev/api-docs/llm.rb/LLM/Stream.html) with structured callbacks for content,
+reasoning, tool calls, tool returns, and compaction.
 
 ```ruby
-require "llm"
-
-class Stream < LLM::Stream
+class MyStream < LLM::Stream
   def on_content(content)
-    $stdout << content
+    print content
   end
-end
 
-llm = LLM.openai(key: ENV["KEY"])
-agent = LLM::Agent.new(llm, stream: Stream.new)
-agent.talk "Write a haiku about Ruby."
-```
-
-#### LLM::Stream (advanced)
-
-The
-[LLM::Stream](https://0x1eef.github.io/x/llm.rb/LLM/Stream.html)
-object can also resolve tool calls while output is still streaming. In
-`on_tool_call`, you can spawn the tool, push the work onto the stream
-queue, and later drain it with `wait`:
-
-```ruby
-require "llm"
-
-class Stream < LLM::Stream
-  def on_content(content)
-    $stdout << content
-  end
-
-  def on_tool_call(tool, error)
-    return queue << error if error
-    queue << ctx.spawn(tool, :thread)
+  def on_reasoning_content(content)
+    warn content
   end
 end
 
-llm = LLM.openai(key: ENV["KEY"])
-ctx = LLM::Context.new(llm, stream: Stream.new, tools: [ReadFile])
-ctx.talk "Read README.md and summarize the quick start."
-ctx.talk(ctx.wait) while ctx.functions?
-```
-
-#### Concurrency
-
-llm.rb can run tool work concurrently. This is useful when a model calls
-multiple tools and you want to resolve them in parallel instead of one
-at a time. On
-[LLM::Agent](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html),
-you can enable this with `concurrency`. Common options are `:call` for
-sequential execution, `:thread`, or `:task` for concurrent IO-bound work, and
-`:ractor` or `:fork` for more isolated CPU-bound work:
-
-```ruby
-require "llm"
-
-class Agent < LLM::Agent
-  model "gpt-5.4-mini"
-  tools ReadFile
-  concurrency :thread
-end
-
-llm = LLM.openai(key: ENV["KEY"])
-agent = Agent.new(llm, stream: $stdout)
-agent.talk "Read README.md and CHANGELOG.md and compare them."
-```
-
-#### Serialization
-
-The [`LLM::Agent`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html)
-object can be serialized to JSON, which makes it suitable for storing
-in a file, a database column, or a Redis queue. The built-in
-ActiveRecord and Sequel plugins are built on top of the same underlying
-serialization feature:
-
-```ruby
-require "llm"
-
-llm = LLM.openai(key: ENV["KEY"])
-
-# Serialize an agent
-agent1 = LLM::Agent.new(llm)
-agent1.talk "Remember that my favorite language is Ruby"
-string = agent1.to_json
-
-# Restore an agent (from JSON)
-agent2 = LLM::Agent.new(llm, stream: $stdout)
-agent2.restore(string:)
-agent2.talk "What is my favorite language?"
+llm = LLM.deepseek(key: ENV["KEY"])
+agent = LLM::Agent.new(llm, stream: MyStream.new)
+agent.talk "Explain Ruby fibers."
 ```
 
-#### ask
+#### LLM::MCP
 
-[`LLM::Agent`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html)
-also provides `ask`, a convenience interface that is compatible with
-RubyLLM's `ask` method. It accepts a prompt, an optional `with:`
-attachment path or paths, an optional `stream:` target, and an optional
-block that chunks are yielded to. It returns an
-[`LLM::Response`](https://0x1eef.github.io/x/llm.rb/LLM/Response.html),
-so use `.content` when you want the text directly:
+The Model Context Protocol (MCP) has first-class support
+in llm.rb. The stdio and http transports work out of the
+box. MCP tools are translated into subclasses of
+[`LLM::Tool`](https://r.uby.dev/api-docs/llm.rb/LLM/Tool.html) that can be used with [`LLM::Context`](https://r.uby.dev/api-docs/llm.rb/LLM/Context.html)
+or [`LLM::Agent`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html).
 
 ```ruby
 require "llm"
 
-llm = LLM.openai(key: ENV["KEY"])
-agent = LLM::Agent.new(llm)
-
-puts agent.ask("Hello world").content
-puts agent.ask("Summarize this document.", with: "README.md").content
-agent.ask("Stream this reply.") { $stdout << _1 }
-```
-
-## Installation
-
-```bash
-gem install llm.rb
+llm   = LLM.deepseek(key: ENV["KEY"])
+mcp   = LLM::MCP.stdio(argv: ["ruby", "server.rb"])
+agent = LLM::Agent.new(llm, stream: $stdout, tools: mcp.tools)
+agent.talk "Run the tool"
 ```
 
-## Examples
-
-#### REPL
+#### LLM::A2A
 
-This example uses [`LLM::Agent`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html)
-for an interactive REPL. <br> See the
-[deepdive (web)](https://llmrb.github.io/llm.rb/) or
-[deepdive (markdown)](resources/deepdive.md) for more examples.
+The Agent 2 Agent (A2A) protocol has first-class support
+in llm.rb. The http and jsonrpc transports work out of the
+box. A2A skills are translated into subclasses of
+[`LLM::Tool`](https://r.uby.dev/api-docs/llm.rb/LLM/Tool.html) that can be used with [`LLM::Context`](https://r.uby.dev/api-docs/llm.rb/LLM/Context.html)
+or [`LLM::Agent`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html).
 
 ```ruby
 require "llm"
 
-llm = LLM.openai(key: ENV["KEY"])
-agent = LLM::Agent.new(llm, stream: $stdout)
-
-loop do
-  print "> "
-  agent.talk(STDIN.gets || break)
-  puts
-end
+llm   = LLM.deepseek(key: ENV["KEY"])
+a2a   = LLM::A2A.rest(url: "https://remote-agent.example.com")
+agent = LLM::Agent.new(llm, stream: $stdout, tools: a2a.skills)
+agent.talk "Run the skill"
 ```
 
-#### Multimodal: Local Files
+#### RAG
 
-In llm.rb, a prompt can be a string, an [`LLM::Prompt`](https://0x1eef.github.io/x/llm.rb/LLM/Prompt.html), or an array.
-When you use an array, each element can be plain text or a tagged object such as
-[`agent.image_url(...)`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html#image_url-instance_method),
-[`agent.local_file(...)`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html#local_file-instance_method),
-or [`agent.remote_file(...)`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html#remote_file-instance_method).
-Those tagged objects carry the metadata the provider adapter needs to turn one
-Ruby prompt into the provider-specific multimodal request schema.
+Most providers offer an embedding model that can be
+used for semantic search, or similarity search. An
+embedding model can generate embeddings that can then
+be stored in a database that is optimized for storing
+and querying vectors, such as SQLite's [sqlite-vec](https://github.com/asg017/sqlite-vec)
+or PostgreSQL's [pg-vector](https://github.com/pgvector/pgvector).
 
-If the model understands that file type, you can attach a local file directly
-with `agent.ask(..., with: path)` instead of uploading it first through a
-provider Files API. Under the hood, llm.rb tags the path as a
-[`agent.local_file(...)`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html#local_file-instance_method)
-object:
+llm.rb also includes support for OpenAI's vector store API. It
+provides a vector database as a HTTP service but we won't cover
+that here.
 
 ```ruby
 require "llm"
 
-llm = LLM.openai(key: ENV["KEY"])
-agent = LLM::Agent.new(llm)
-puts agent.ask("Summarize this document.", with: "README.md").content
-```
-
-#### Context Compaction
-
-This example uses [`LLM::Agent`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html),
-[`LLM::Compactor`](https://0x1eef.github.io/x/llm.rb/LLM/Compactor.html), and
-[`LLM::Stream`](https://0x1eef.github.io/x/llm.rb/LLM/Stream.html) together so
-long-lived conversations can summarize older history and expose the lifecycle
-through stream hooks. This approach is inspired by General Intelligence
-Systems. The
-compactor can also use its own `model:` if you want summarization to run on a
-different model from the main conversation. `token_threshold:` accepts either a
-fixed token count or a percentage string like `"90%"`, which resolves
-against the active model context window and triggers compaction once total
-token usage goes over that percentage. See the
-[deepdive (web)](https://llmrb.github.io/llm.rb/) or
-[deepdive (markdown)](resources/deepdive.md) for more examples.
-
-```ruby
-require "llm"
+llm  = LLM.openai(key: ENV["KEY"])
+body = "llm.rb is Ruby's capable AI runtime."
+embedding = llm.embed([body]).embeddings.first
 
-class Stream < LLM::Stream
-  def on_compaction(ctx, compactor)
-    puts "Compacting #{ctx.messages.size} messages..."
-  end
-
-  def on_compaction_finish(ctx, compactor)
-    puts "Compacted to #{ctx.messages.size} messages."
-  end
-end
-
-llm = LLM.openai(key: ENV["KEY"])
-agent = LLM::Agent.new(
-  llm,
-  stream: Stream.new,
-  compactor: {
-    token_threshold: "90%",
-    retention_window: 8,
-    model: "gpt-5.4-mini"
-  }
+Document.create!(
+  title: "llm.rb",
+  body:,
+  embedding:,
 )
 ```
 
-#### Reasoning
-
-This example uses [`LLM::Stream`](https://0x1eef.github.io/x/llm.rb/LLM/Stream.html)
-with the OpenAI Responses API so reasoning output is streamed separately from
-visible assistant output. See the
-[deepdive (web)](https://llmrb.github.io/llm.rb/) or
-[deepdive (markdown)](resources/deepdive.md) for more examples.
-
-To use the Responses API (OpenAI-specific), initialize an agent with
-`mode: :responses` and keep using `talk` for turns.
-
-```ruby
-require "llm"
-
-class Stream < LLM::Stream
-  def on_content(content)
-    $stdout << content
-  end
-
-  def on_reasoning_content(content)
-    $stderr << content
-  end
-end
-
-llm = LLM.openai(key: ENV["KEY"])
-agent = LLM::Agent.new(
-  llm,
-  model: "gpt-5.4-mini",
-  mode: :responses,
-  reasoning: {effort: "medium"},
-  stream: Stream.new
-)
-agent.talk("Solve 17 * 19 and show your work.")
-```
-
-#### Request Cancellation
-
-Need to cancel a stream? llm.rb has you covered through
-[`LLM::Agent#interrupt!`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html#interrupt-21-instance_method).
-<br> See the [deepdive (web)](https://llmrb.github.io/llm.rb/)
-or [deepdive (markdown)](resources/deepdive.md) for more examples.
-
-```ruby
-require "llm"
-require "io/console"
-
-llm = LLM.openai(key: ENV["KEY"])
-agent = LLM::Agent.new(llm, stream: $stdout)
-worker = Thread.new do
-  agent.talk("Write a very long essay about network protocols.")
-rescue LLM::Interrupt
-  puts "Request was interrupted!"
-end
-
-STDIN.getch
-agent.interrupt!
-worker.join
-```
+#### Concurrency
 
-#### Sequel (ORM)
+The runtime supports five different concurrency strategies that have
+different attributes. The choice between all of them often depends
+on the requirements of your application.
 
-The `plugin :llm` integration wraps
-[`LLM::Context`](https://0x1eef.github.io/x/llm.rb/LLM/Context.html) on a
-`Sequel::Model` and keeps tool execution explicit. Like the ActiveRecord
-wrappers, its built-in persistence contract is the serialized `data` column,
-while `provider:` resolves a real `LLM::Provider` instance and `context:`
-injects defaults such as `model:`. <br> See the
-[deepdive (web)](https://llmrb.github.io/llm.rb/) or
-[deepdive (markdown)](resources/deepdive.md) for more examples.
+IO-bound tools are a good fit for the `:task`, `:thread`,
+and `:fiber` strategies while true parallelism can be achieved
+with the `:fork` and `:ractor` strategies. The
+`:fork` strategy also provides a separate process that offers
+isolation from its parent.
 
 ```ruby
 require "llm"
-require "net/http/persistent"
-require "sequel"
-require "sequel/plugins/llm"
-
-class Context < Sequel::Model
-  plugin :llm, provider: :set_provider, context: :set_context
-
-  private
 
-  def set_provider
-    LLM.openai(key: ENV["OPENAI_SECRET"], persistent: true)
-  end
-
-  def set_context
-    {model: "gpt-5.4-mini", mode: :responses, store: false}
-  end
-end
-
-ctx = Context.create
-ctx.talk("Remember that my favorite language is Ruby")
-puts ctx.talk("What is my favorite language?").content
+llm   = LLM.deepseek(key: ENV["KEY"])
+tools = [FetchNews, FetchStocks, FetchFeeds]
+agent = LLM::Agent.new(llm, tools:, concurrency: :fork)
+agent.talk "Run the tools in parallel"
 ```
 
-#### ActiveRecord (ORM): acts_as_llm
+#### ORM
 
-The `acts_as_llm` method wraps [`LLM::Context`](https://0x1eef.github.io/x/llm.rb/LLM/Context.html) and
-provides full control over tool execution. Its built-in persistence contract is
-one serialized `data` column. If your app has provider, model, or usage
-columns, provide them to llm.rb through `provider:` and `context:` instead of
-relying on reserved wrapper columns.
+Because both [`LLM::Context`](https://r.uby.dev/api-docs/llm.rb/LLM/Context.html), and [`LLM::Agent`](https://r.uby.dev/api-docs/llm.rb/LLM/Agent.html)
+can be serialized to JSON and stored in a simple string, both ActiveRecord
+and Sequel support can be implemented within a single column on a single row.
 
-See the [deepdive (web)](https://llmrb.github.io/llm.rb/)
-or [deepdive (markdown)](resources/deepdive.md) for more examples.
+The runtime includes first-class support for both ActiveRecord *and* Sequel, and
+for both Rack-based applications *and* Rails-based applications. On databases
+where it is supported, such as PostgreSQL, the column can be optimized by using
+the `jsonb` type.
 
 ```ruby
-require "llm"
 require "active_record"
-require "llm/active_record"
-
-class Context < ApplicationRecord
-  acts_as_llm provider: :set_provider, context: :set_context
-
-  private
-
-  def set_provider
-    LLM.openai(key: ENV["OPENAI_SECRET"])
-  end
-
-  def set_context
-    {model: "gpt-5.4-mini", mode: :responses, store: false}
-  end
-end
-
-ctx = Context.create!
-ctx.talk("Remember that my favorite language is Ruby")
-puts ctx.talk("What is my favorite language?").content
-```
-
-```ruby
 require "llm"
-require "active_record"
 require "llm/active_record"
 
-class Context < ApplicationRecord
-  acts_as_llm provider: :set_provider, context: :set_context
-
-  # Optional application columns can still provide the provider and context.
-  # For example, `provider_name` and `model_name` can be normal columns.
-
-  private
-
-  def set_provider
-    LLM.public_send(provider_name, key: provider_key)
-  end
-
-  def set_context
-    {model: model_name, mode: :responses, store: false}
+class Agent < ApplicationRecord
+  acts_as_agent do |agent|
+    agent.model "deepseek-v4-pro"
+    agent.instructions "solve the user's query"
+    agent.tools [Research, FinalizeResearch, ActOnResearch]
   end
-end
-```
-
-#### ActiveRecord (ORM): acts_as_agent
-
-The `acts_as_agent` method wraps [`LLM::Agent`](https://0x1eef.github.io/x/llm.rb/LLM/Agent.html) and
-manages tool execution for you. Like `acts_as_llm`, its built-in persistence
-contract is one serialized `data` column. If your app has provider or model
-columns, provide them to llm.rb through your hooks and agent DSL.
-
-See the [deepdive (web)](https://llmrb.github.io/llm.rb/)
-or [deepdive (markdown)](resources/deepdive.md) for more examples.
-
-```ruby
-require "llm"
-require "active_record"
-require "llm/active_record"
-
-class Ticket < ApplicationRecord
-  acts_as_agent provider: :set_provider, context: :set_context
-  model "gpt-5.4-mini"
-  instructions "You are a concise support assistant."
-  tools SearchDocs, Escalate
-  concurrency :thread
 
   private
 
+  # By convention, this method defines the provider for a model.
+  # If necessary, it can be renamed with: provider: :your_method.
   def set_provider
-    LLM.openai(key: ENV["OPENAI_SECRET"])
+    LLM.deepseek(key: ENV["KEY"])
   end
 
+  # By convention, this method returns the context options given
+  # to LLM::Context or LLM::Agent.
   def set_context
-    {mode: :responses, store: false}
+    {}
   end
 end
 
-ticket = Ticket.create!
-puts ticket.talk("How do I rotate my API key?").content
+agent = Agent.create!
+agent.talk "perform research"
 ```
 
-```ruby
-require "llm"
-require "active_record"
-require "llm/active_record"
+## FAQ
 
-class Ticket < ApplicationRecord
-  acts_as_agent provider: :set_provider, context: :set_context
-  model "gpt-5.4-mini"
-  instructions "You are a concise support assistant."
+<details>
+<summary>What providers does llm.rb support?</summary>
+<br>
+<p>
+China-based
 
-  private
+* DeepSeek
+* zAI
 
-  def set_provider
-    LLM.public_send(provider_name, key: provider_key)
-  end
+US-based
 
-  def set_context
-    {mode: :responses, store: false}
-  end
-end
-```
+* OpenAI
+* Google (Gemini)
+* xAI
+* AWS bedrock
+* DeepInfra
+* Anthropic
 
-#### MCP
+Openweights
 
-This example uses [`LLM::MCP`](https://0x1eef.github.io/x/llm.rb/LLM/MCP.html)
-over HTTP so remote GitHub MCP tools run through the same
-`LLM::Agent` tool path as local tools. It expects a GitHub token in
-`ENV["GITHUB_PAT"]`. See the
-[deepdive (web)](https://llmrb.github.io/llm.rb/) or
-[deepdive (markdown)](resources/deepdive.md) for more examples.
+* DeepSeek
+* zAI
+* DeepInfra
+* AWS bedrock
 
-```ruby
-require "llm"
-require "net/http/persistent"
+Host your own
 
-llm = LLM.openai(key: ENV["KEY"], persistent: true)
-mcp = LLM::MCP.http(
-  url: "https://api.githubcopilot.com/mcp/",
-  headers: {"Authorization" => "Bearer #{ENV["GITHUB_PAT"]}"},
-  persistent: true
-)
+* Ollama
+* Llamacpp
+</p>
+</details>
 
-agent = LLM::Agent.new(llm, stream: $stdout, tools: mcp.tools)
-agent.talk("Pull information about my GitHub account.")
-```
+<details>
+<summary>I have a limited budget. What should I do?</summary>
+<br>
+<p>
+There a few options. The first option is to host
+your own model, and use the ollama or llamacpp
+providers. This can be diffilcult though because
+a capable model requires hardware that can
+match it. If you have the ability to self-host,
+this would be my first option.
+</p>
+<p>
+The second option is DeepSeek. <br>
+The deepseek-v4-flash model costs pennies to use. <br>
+And llm.rb has been optimized for deepseek. For example,
+DeepSeek does not have image generation capabilities
+but on the llm.rb runtime it does (vector graphics only,
+though).
+</p>
+<p>
+The same is true for structured outputs. DeepSeek does
+not support structured outputs in the same way as OpenAI or
+Google, but the llm.rb runtime makes it appear as
+though it does, through the `json_object` response
+type.
+</p>
+If you're on a budget, DeepSeek is hard to beat.
+</details>
+<details>
+<summary>Can I download llm.rb via a decentralized network?</summary>
+<br>
+You can!
+<br>
+We are on the <a href="https://radicle.network">radicle.network</a>
+<br>
+Every commit that lands on GitHub also lands on Radicle.
+<br>
+Our repository ID is z2PtfQ6dYwyYaW2aGrztG1sMyDmCE.
+<br>
+Browse on <a href="https://radicle.network/nodes/iris.radicle.network/z2PtfQ6dYwyYaW2aGrztG1sMyDmCE">the web</a>.
+</details>
 
 ## Resources
 
-- [deepdive (web)](https://llmrb.github.io/llm.rb/) and
-  [deepdive (markdown)](resources/deepdive.md) are the examples guide.
-- [relay](https://github.com/llmrb/relay) shows a real application built on
-  top of llm.rb.
-- [doc site](https://0x1eef.github.io/x/llm.rb?rebuild=1) has the API docs.
+If you like what you read so far, check out the [deepdive.md](https://r.uby.dev/llm/deepdive/)
+to learn more. Unfortunately it
+wasn't possible to cover every feature without the README becoming a small book.
+The [r.uby.dev](https://r.uby.dev) homepage also includes more learning material
+and resources.
 
 ## License
 
-[BSD Zero Clause](https://choosealicense.com/licenses/0bsd/)
+[Business Source License 1.1](./LICENSE)
+<br>
+Commercial production use requires a commercial license.
+<br>
+Each version converts to the [BSD Zero Clause](https://choosealicense.com/licenses/0bsd/)
+four years after its first public release.
 <br>
-See [LICENSE](./LICENSE)
+Contact [robert@r.uby.dev](mailto:robert@r.uby.dev) for a commercial license.