Opus, Sonnet, and Haiku behind a single Anthropic-compatible endpoint. Point your base URL at us and keep every line of your code exactly where it is.
# one line changes — the rest of your code doesn't export ANTHROPIC_BASE_URL="https://claude.mrvinci.app" export ANTHROPIC_AUTH_TOKEN="sk_mv_…"
Ask for any model in the lineup by id — see /v1/models for the live list.
Flagship reasoning and long-horizon agentic work.
high effortPrior flagship — deep reasoning, still formidable.
high effortCapable Opus tier for demanding tasks.
high effortThe sweet spot of speed and intelligence.
high effortFast, balanced everyday model.
medium effortReliable workhorse for most tasks.
medium effortLowest latency, highest throughput.
low effortYour agent talks to one endpoint. Every request is metered against your key and the answer streams straight back — any model in the lineup, same key.
Mint a key per team or per app. Revoke or top up any one without touching the rest.
A 5-hour usage window (anchored at your first message, like Claude) plus a 7-day cap, with optional per-minute limits — enforced per key. Keys can carry an expiry date.
Server-sent events with no added latency — text, tools, and thinking blocks stream exactly as your SDK expects.
Send screenshots and images exactly as the API specifies — vision requests just work.
Works with the official SDKs and with Claude Code, Claude Desktop, Cursor, and Cline — only the base URL changes.
Every key has a status page: spend or window consumption, per-model breakdown, and request counts.
Ask an admin for an sk_mv_ key, or mint one in the MrVinci admin panel.
Run npx mrvinciapi and paste the key, or set ANTHROPIC_BASE_URL yourself.
Start a new session — your traffic now flows through the gateway. No code changes.