Documentation

Zenn.Ceo API

Claude models, image generation and video generation behind one ck_ API key. The API is invite-only and every account is locked to its own IP / domain allowlist.

API access (invite-only)

Since November 2, 2026 API keys only work for approved accounts, and only from the IP addresses or domains registered for that account. Online payments are disabled; credits are allocated by the team (1,000,000 credits = $1).

  1. Create an account at /auth/register and verify your email.
  2. Email support@zenn.ceo from that address with your use case, expected volume, and the public egress IPs / CIDR ranges and/or domains your requests come from.
  3. After approval, create a key at /manage-api-keys and call the API from your server.
  • IPs: exact IPv4/IPv6 or CIDR (IPv4 /8–/32, IPv6 /32–/128). The client IP is the address that connects to us; X-Forwarded-For you send is ignored.
  • Domains: we resolve A/AAAA records (cached up to 5 minutes) and accept those IPs. Use a hostname that points at your outbound server, not a CDN.
  • All keys on an account share its allowlist. Changes apply within about a minute.
  • Only the endpoints on this page exist. Requests to any other path, without a key, or with an invalid key can get the source IP blocked permanently.

Base URL & authentication

Base URL: https://api.zenn.ceo/v1

The API is served only on api.zenn.ceo. Send your key in one of these headers:

HeaderValue
AuthorizationBearer ck_...
x-api-keyck_...

Claude Code

Claude Code appends /v1/messages itself, so use the origin:

shell
export ANTHROPIC_BASE_URL=https://api.zenn.ceo
export ANTHROPIC_AUTH_TOKEN=ck_YOUR_API_KEY
claude

Any Anthropic SDK works the same way with base_url="https://api.zenn.ceo".

POST /v1/messages

Anthropic Messages API, including streaming and tool use.

cURL
curl https://api.zenn.ceo/v1/messages \
  -H "x-api-key: $ZENN_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4-6",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello"}]
  }'
ModelInput / MTokOutput / MTok
claude-opus-4-7$15.00$75.00
claude-opus-4-6$15.00$75.00
claude-sonnet-4-6$3.00$15.00
claude-haiku-4-5$1.00$5.00

POST /v1/images/generations

OpenAI-compatible image generation (synchronous).

cURL
curl https://api.zenn.ceo/v1/images/generations \
  -H "Authorization: Bearer $ZENN_API_KEY" \
  -H "content-type: application/json" \
  -d '{"model": "gpt-image-2", "prompt": "a red fox in snow", "n": 1}'

POST /v1/images

Image models from GET /v1/models with endpoint: "/v1/images" (text-to-image, image edit, face swap, upscale). Returns the result directly, or 202 with a pollUrl for slower jobs.

cURL
curl https://api.zenn.ceo/v1/images \
  -H "Authorization: Bearer $ZENN_API_KEY" \
  -H "content-type: application/json" \
  -d '{"model": "MODEL_ID", "input": {"prompt": "a red fox in snow"}}'

# → {"status":"completed","outputUrl":"https://...","jobId":"...","creditsUsed":30000}
# or 202 {"status":"processing","jobId":"...","pollUrl":"/v1/jobs?id=...&model=...&type=image"}

POST /v1/videos

Video models from GET /v1/models with endpoint: "/v1/videos". Always asynchronous: poll the returned pollUrl.

cURL
curl https://api.zenn.ceo/v1/videos \
  -H "Authorization: Bearer $ZENN_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "MODEL_ID",
    "input": {"prompt": "slow dolly shot", "image": "https://...", "duration": 5}
  }'

# → 202 {"status":"processing","jobId":"...","pollUrl":"https://api.zenn.ceo/v1/jobs?id=..."}

GET /v1/jobs

Poll a job using the exact pollUrl returned by /v1/images or /v1/videos, every 3–5 seconds, until status is completed or failed.

cURL
curl "https://api.zenn.ceo/v1/jobs?id=JOB_ID&model=MODEL_ID&type=video" \
  -H "Authorization: Bearer $ZENN_API_KEY"

# → {"status":"processing"}
# → {"status":"completed","outputUrl":"https://...","jobId":"JOB_ID"}
# → {"status":"failed","error":"..."}

GET /v1/models · GET /v1/credits

cURL
curl https://api.zenn.ceo/v1/models -H "Authorization: Bearer $ZENN_API_KEY"
# → {"object":"list","data":[{"id":"...","type":"video","endpoint":"/v1/videos","pricing":{...}}, ...]}

curl https://api.zenn.ceo/v1/credits -H "Authorization: Bearer $ZENN_API_KEY"
# → {"object":"credit_balance","available_credits":12500000}

Errors & limits

{"error": {"type": "permission_error", "code": "ip_not_allowed", "message": "..."}}
StatusMeaning
400Invalid request (missing model, bad input)
401Missing or invalid API key
402Insufficient credits — contact support
403api_access_not_enabled account not approved · ip_not_allowed request IP not on the allowlist
429Rate limit or per-key spending limit reached
503allowlist_unavailable or upstream unavailable — retry with backoff

Rate limits are per account (about 1,000 requests/hour for messages, lower for media). Each key can also have a monthly spending limit set at /manage-api-keys.