Read the gateway like a manual, not a maze.
Quick start, API reference, guides, and architecture — grouped so you can jump straight to the part you need, with pricing that stays predictable in the currency you already use.
Start here
Quick start
08Overview
What Relixr is and how requests flow through it.
Create your first API key
Get an inference key from the dashboard.
Make a request
OpenAI-compatible gateway endpoint.
SDKs
JavaScript and Python packages (separate repos).
CLI
relixr login, doctor, keys, agents demo.
Framework guides
LangChain, Vercel AI SDK, LiteLLM, CrewAI.
Supported models
Anthropic, OpenAI, Google, Mistral, and more.
OpenAI compatibility
What Relixr supports — and what it does not.
API reference
API reference
08Chat completions
POST /v1/chat/completions
Embeddings
POST /v1/embeddings
Authentication
Inference vs management keys.
Errors
Status codes and the gateway error contract.
OpenAPI
OpenAPI 3 spec and try-it guidance.
Management API
Keys, agents, usage, and balance APIs.
Headers & metadata
Request id, cost, and routing fields.
Deprecation policy
How model ids and APIs are retired.
Guides
Features
06Model routing
Declarative routing rules and transparency.
Guardrails
PII, injection detection, geo, spend limits.
Bring your own keys
Use your provider keys with Relixr routing.
Presets
Reusable prompt + parameter bundles.
Agents
Fleet identity, budgets, and key binding.
Billing & credits
Top-ups, balance reserve, and cost breakdown.
Design guides
02Designing cost-aware UX
Show users what a request costs before and after it runs, without cluttering the UI.
Streaming and loading states
UI patterns for partial tokens, mid-stream failures, retries, and cancel.
Product guides
02Choosing a model for a task
A practical framework for matching a job to a model family and price point.
Setting spend controls
Budgets, alerts, and hard caps for teams and agents.
Feature guides
03API key scopes
Restrict a key to specific models, routes, or spend before you hand it out.
Usage analytics
Read the activity timeline, volume chart, and per-key cost breakdown.
Roles, audit log, and model governance
The permission taxonomy behind roles, the audit trail, and the model request/approve flow.
Persona guides
05For solo developers
One key, sane defaults, and a bill you can predict.
For lean and budget-conscious teams
Predictable, prepaid pricing with no subscription commitment and no international card required.
For platform teams
Multi-project keys, shared budgets, and audit trails.
For agent builders
Fleet identity, per-agent budgets, and attribution.
For enterprise teams
Roles, audit trails, model governance, and budget controls for organizations that need to prove who did what.
Cookbooks
03Unstick Cursor in 5 minutes
Jam-recovery when the subscription meter runs out.
Agent sandbox 0→1
Create, demo, prove cost and attribution.
Rolling out to an enterprise workspace
Roles, budgets, guardrails, and governance, in the order to set them up.
Architecture
Reference architecture
03Agent fleet on Relixr
A reference layout for running many autonomous agents behind one gateway.
Multi-provider failover
Routing rules that survive a provider outage without a code change.
Guardrails as a perimeter
Where PII, injection, geo, and spend checks sit relative to your app.
Migrate & connect
Migrations
04From OpenAI
Point the OpenAI SDK at Relixr.
From OpenRouter
Community ids and prepaid credits.
From Anthropic-only
Claude plus multi-provider failover.
nf_live_ → rlx_
Key prefix migration window.
Integrations
04Use Relixr with Cursor
Point Cursor at Relixr for uncapped model choice and hard spend caps.
Use Relixr with Claude Code
Swap ANTHROPIC_BASE_URL to run Claude Code through Relixr.
Use Relixr with VS Code
Continue, Cline, Roo Code, and Copilot BYOK, configured for Relixr.
Use Relixr with other tools
Aider, Codex CLI, Zed, Neovim, and automation platforms like n8n.