Everything you need to run Claude Code
Our LLM over an Anthropic-compatible API — built for Claude Code and the Anthropic SDK.
Fast API
Low-latency edge routing and connection reuse keep round-trips snappy under load.
Built for Claude Code
Set two env vars and Claude Code talks to our LLM — no SDK changes, no config rewrites.
Streaming Support
Native server-sent events stream tokens back the moment the model produces them.
API Key Management
Create, rename, disable, and revoke scoped keys. Only hashes are ever stored.
Usage Analytics
Track requests, latency, and token counts per key with real-time dashboards.
High Availability
Self-healing failover routes around rate limits and errors automatically — zero downtime for you.
Start free in Claude Code, scale when you need to
Every account starts on a free 30-day Pro Trial — no card required. Upgrade any time for higher rate limits.
Frequently asked questions
Set ANTHROPIC_BASE_URL to https://api.lurevixis.online and ANTHROPIC_API_KEY to your vix_ key, then run claude as usual. Claude Code talks to our LLM with no other changes — it works with the Anthropic SDK the same way.
All of them. Request bodies and Anthropic headers are forwarded transparently, so streaming SSE, tool use, image inputs, system prompts, and extended thinking all work in Claude Code — including future Anthropic fields.
Keys follow the format vix_ + 16 characters and are shown to you exactly once. Only a salted SHA-256 hash is stored — we can never reveal a key after creation.
Yes. Every new account starts on a 30-day Pro Trial — no card required. Create a key, plug it into Claude Code, and start building right away. Upgrade any time for higher rate limits.
Self-healing failover handles it. If capacity is exhausted, rate limited, or errors out, requests automatically retry on the next available path with zero downtime for you.
Your dashboard shows total requests, requests today, active keys, total token usage, and a request graph — plus per-request logs with latency and status.