Tool calling
Its arguments. The function cannot know who called it, what else is running, or what happens next.
Ship apps that think and act.
Build your harness in TypeScript. Ship it as a single on-device binary that runs offline, as a multi-user service on a private appliance, or online on frontier compute.
$
npx lloyal-ai newRecognised by
NVIDIA Inception program member
Endorsed publicly by RunPod's CEO
lloyal-aiguided application scaffold
No OpenAI or Anthropic key. No Docker, no Ollama, no LangGraph, no vector database. Just your code and the model. Read the ten-minute guide →
01 / ABILITIES
Extend. Interact. Recurse.
Agents use Abilities to act on the world and on live inference simultaneously.
Abilities are inference-native. They run inside the model’s execution rather than behind a network boundary, so an extension can see who called it, what that caller already knows, and what else is running — and can fork that state into new agents.
Lloyal hosts a signed channel of 1st & 3rd party Abilities. Installing one brings it all into your app in a single command, for your in-app agents to consume: tools and their schemas, a live source, situation-aware skills[1], and the models it composes. DeepSeek’s Harness bets on a plugin ecosystem for its own agent. Abilities are that ecosystem for the apps you ship.
[1] Agent count · Other agents’ tasks · Tool budgets · Chain position · Date
How Abilities work →Browse Abilities →MCP
Its arguments, in its own process. Language freedom and isolation — the right default for most integrations.
Skills
Its arguments, and the context window. The extension carries procedure now, not only operations.
Abilities
The execution itself. The calling agent, its live inference state, its lineage, its siblings — the authority to fork it into new agents, and to require the harness to load another model before it will run.
02 / Architecture
Vertical
Inference
System intelligence beyond the model.
Vertical Inference is the architecture in which the application, the model runtime and live inference state become one programmable system. Ordinary code governs how intelligence retrieves, reasons, acts and continues while it is underway.
Lloyal is the platform for building and deploying this architecture—from local applications to frontier compute.
Inspect a working TypeScript application ↗From the engineering blog
Harness Development Kit
Ship local AI apps in TypeScript — one command scaffolds a working app with the model, agents, retrieval and runtime already inside it.
Agents that share memory on the GPU
Down to the attention mask: how a fork becomes an owner bit per cell, and a cohort becomes one batched decode.
You already know this architecture
It shipped in 2007. Lloyal’s serving stack maps noun for noun onto the stack that carried Rails — Rack→binding, Puma→host, Unicorn→relay, config.ru→driver.
Shifting the harness left
Ten concurrent GLM-5.2 agents for the compute of one — one shared KV context, one decode per step, paid in memory not compute. Full trace and cost receipt published.
All dispatches →
03 / THE HARNESS
The Model in MVC now thinks.
For two decades the Model in MVC was inert—rows in a database, waiting to be queried. Put a live language model in that slot and every other part holds: product surfaces are still the Views, and the harness is the Controller, ordinary TypeScript deciding how the application's intelligence collaborates, acts, recovers and continues. What changes is that this Model thinks, and it keeps thinking while your code governs it—the same code whether it runs a 4B model on a laptop or GLM-5.2 across a cluster of B200s.
Views
CLI · Desktop · Web · Mobileproduct surfaces · execution environments
Controller
Your harnesstypescript · policy · capabilities · control flow
Model
Resident LLMshared context · live branches · generative state
The model supplies the intelligence. The harness supplies the institution.
Program the behaviour once — regardless of product surface or compute:
model
Live inference
controller
incident-harness.tsTypeScript + Effection
export function* incidentHarness(incident) { const pool = yield* agentPool({ orchestrate: parallel([ inspectTelemetry(incident), searchServiceHistory(incident), reviewTechnicalManuals(incident), ]), }); const evidence = pool.agents.map((a) => a.result); const assessment = yield* reconcile(evidence); if (assessment.requiresApproval) { return yield* requestOperatorDecision(assessment); } return yield* proposeRemediation(assessment);}view
Application surfaces · execution environments
- incident-4417CLITerminal / TUI
approve remediation?
[y] yes [n] hold
- operatorDesktopElectron / Tauri
Approve remediation
ApproveHold - host.myapp.comWebBrowser
Approve remediation
ApproveHold - Mobile sooniOS / Android · JSI
Approve remediation
Approve
CONTACT
Get in touch.
The platform is public and the docs are open — everything above is self-serve. This is the door for partnerships, the hardware programme, press, and anything the documentation does not answer.
Private AbilitiesYour own signed channelOne program, every placement
EDGEQwen 3.5 · 4B
on-device
4B → APP
planning · agents
tools · recovery · synthesis
tools · recovery · synthesis
ONE APPLICATION
preflight()
plan()
agentPool()
policy()
synthesize()
commitTurn()
agent:spawn
same program · same Abilities · same event contract
FRONTIERGLM-5.2 · 2×B200
hosted
10 → 1
ten live agents
one model dispatch per step
one model dispatch per step
Two real runs. One flagship application built on Lloyal, spanning both extremes.
START
All of it ships from one command.
$
npx lloyal-ai newNo API key, no inference server, no vector database. Ten-minute guide · Docs · HDK on GitHub