the wall of models
eisencode,
one endpoint.
12 open models behind one OpenAI-style API. sign up, get a key on the spot, call from anywhere — opencode, cursor, curl, anything. no access without your key. your credits, your requests, your log.
12 modelsopenai-stylekey-gatedcredit metered
SEC.01 — the catalog
twelve names. one base url.
eisenberg/deepseek-v4.1-flash
eisenberg/flint-40b
eisenberg/glm-5.3-flash
eisenberg/beam
eisenberg/big-pickle
eisenberg/iceman
eisenberg/inkling-2-small
eisenberg/longcat-2.5
eisenberg/mimo-v2.6-flash
eisenberg/minimax-m3.1-flash
eisenberg/nemotron-3.5-lightning
eisenberg/nemotron-3.5-super-vl
SEC.02 — the call
works where you already work.
curl https://eisencode.waivy.dev/api/v1/chat/completions \
-H "Authorization: Bearer sk_eis_..." \
-H "Content-Type: application/json" \
-d '{
"model": "eisenberg/glm-5.3-flash",
"messages": [{"role": "user", "content": "hello"}]
}'point any openai client at the base url — cursor, opencode, cline, the python sdk. the model id is the only thing you change.
SEC.03 — the rules
no key, no answer
every request is checked against a live key. wrong key: 401. no balance: 402. that's the wall.
credits on the wall
sign up and get 25 free credits. issue keys from your wallet, revoke anytime — unused credits refund.
every call is logged
model, latency, status — your dashboard keeps the ledger of what you actually used.