AGENT-NATIVE AI INFRASTRUCTURE

Your agents think in goals. You shouldn't have to think in tokens.

ELTEX Agent Plan gives compatible AI agents recurring multi-model capacity, Smart Routing, and clear usage controls through one monthly subscription.

DEFINITION

What is an agent-native Agent Plan?

An agent-native Agent Plan provides recurring multi-model capacity and runtime infrastructure around autonomous workloads. ELTEX combines routing and usage controls with Agent Refill, durable Agent State, Agent RunGuard safeguards, and Agentic Wallet access for bounded USDC x402 payments.

AGENT INFRASTRUCTURE

Continuity, control, and authorized capacity recovery.

Agent Refill

Allows explicitly authorized agents to activate a stored Reset Ticket when an approved job runs out of capacity.

Agent State

Persists structured progress, decisions, checkpoints, private managed files, and external artifact references across sessions.

Agent RunGuard

Applies versioned budgets, call limits, runtime limits, and loop detection before more work is authorized.

Agentic Wallet

Lets a dedicated agent key request USDC x402 payments within versioned network, amount, domain, approval, pause, and expiry rules.

ELTEX EXPLAINED

A multi-model AI router built around agent workloads.

ELTEX gives compatible AI agents one OpenAI-compatible endpoint for eligible text models, Smart Routing, Preferred and Force controls, recurring plan capacity, Agent Refill, Agent State, Agent RunGuard, Agentic Wallet USDC x402 access, usage records, Reset Tickets, Burst Packs, and separate Elixir media credits.

How ELTEX differs from a direct model API

A direct model API connects an application to one provider. ELTEX adds a routing and capacity layer that can select among configured providers while preserving one application-facing API. Model access and routing remain subject to the live catalog, plan configuration, provider availability, and applicable provider terms.

Frequently asked questions

What is ELTEX Agent Plan?
A recurring-capacity subscription for compatible AI agents with multi-model access, routing controls, usage visibility, and optional Burst protection.
How does Smart Routing work?
ELTEX classifies an eligible request and selects a configured route based on task, complexity, plan access, provider availability, and routing policy.
Is a particular model guaranteed?
No. The live model catalog and plan catalog are the current sources of truth.
Where are API details published?
Connection examples, authentication, model discovery, capacity, and Reset Ticket controls are documented on the API documentation page.