# Ezkielyna API

> AI API gateway at https://api.ezkielyna.store, run by Ezkielyna AI Group (Indonesia). One API key reaches 313 models (238 chat/text models plus image, video and speech) in OpenAI, Anthropic and Gemini request formats. Pay-as-you-go API Credit bought in IDR (QRIS) or USDT at https://ezkielyna.store/api-credit, from IDR 8,000; the balance does not expire. Generated 2026-11-12 02:17 WIB from the live pricing API.

## Quickstart

- Base URL: https://api.ezkielyna.store/v1 (OpenAI SDKs: set base_url to this; Anthropic SDKs and Claude Code: set the base URL to https://api.ezkielyna.store)
- Auth: `Authorization: Bearer <key>` (Anthropic format also accepts `x-api-key: <key>`)
- Get a key: buy API Credit at https://ezkielyna.store/api-credit, then create a key on the API Keys page (https://api.ezkielyna.store/keys). Pick the Auto group so one key covers the Standard, Premium, Hemat and Media groups.
- Test call:

```
curl https://api.ezkielyna.store/v1/chat/completions \
  -H "Authorization: Bearer $EZK_API_KEY" -H "Content-Type: application/json" \
  -d '{"model":"deepseek-v4.1-flash","messages":[{"role":"user","content":"Hello"}]}'
```

- List the models your key can use: GET https://api.ezkielyna.store/v1/models

## Endpoints

- anthropic: POST /v1/messages
- embeddings: POST /v1/embeddings
- gemini: POST /v1beta/models/{model}:generateContent
- image-generation: POST /v1/images/generations
- openai: POST /v1/chat/completions
- openai-audio: POST /v1/audio/transcriptions
- openai-response: POST /v1/responses
- openai-video: POST /v1/videos

## Key groups (a key belongs to one group; the group sets which models it can call and the price multiplier)

- `auto`: Auto (Standard → Premium → Hemat, semua model)
- `default`: Standard 1x (DeepSeek, GLM, Kimi K2, MiniMax, Qwen, MiMo, Hy4 — tanpa Claude/GPT)
- `free`: Hemat 1x (hy3, hy4-preview, hy4-preview-f, edge-tts)
- `media`: Media 1x (gambar & video: GPT Image, Nano Banana, Flux, Imagen, Seedance, Kling, Veo, Wan, Sora, Vidu)
- `premium`: Premium 1.5x (Claude, GPT, Grok, Gemini Pro, Kimi K3, Qwen Max)
- `pro`: Pro 3x (Claude & GPT jalur cadangan — dipakai saat jalur utama penuh)

## Key facts

- Ezkielyna API base URL: https://api.ezkielyna.store/v1 (OpenAI format at /v1/chat/completions, Anthropic format at /v1/messages). Live model list and per-model prices: https://api.ezkielyna.store/pricing
- Key groups: Standard (167 text models, billed at the listed rate; no Claude, GPT, image or video), Premium (91 text models including Claude, GPT and Grok, billed at 1.5x), Hemat (4 budget models: Hunyuan and edge-tts, 1x), Media (75 image and video models, 1x) and Pro (58 models, 3x: Claude, GPT and Grok through a backup route used when the main route is full). A key in the Auto group covers Standard, Premium, Hemat and Media but not Pro, so it is the best default for new users. Image and video need a Media or Auto key; trial and promo accounts cannot create Media keys.
- Pay-as-you-go prices per 1M tokens in IDR at the price of the $1 API Credit pack (IDR 15,000 per $1; the $50 pack works out to IDR 13,300 per $1): deepseek-v4.1-flash input Rp600, cached input Rp81, output Rp2.400; glm-5.3 input Rp3.366, cached input Rp631, output Rp10.519; kimi-k3 input Rp5.185, cached input Rp518, output Rp25.776. Full table: https://ezkielyna.store/harga-api-ai
- Reliability: API Gateway uptime over the last 90 days is 98.47% on the public status page (https://api.ezkielyna.store/status), which also shows per-model status from real requests.
- Ezkielyna is an independent reseller and API gateway. It is not affiliated with OpenAI, Anthropic, Google or xAI.

## Most-used models this week (live price at the group's multiplier)

- deepseek-v4.1-flash-1017 (DeepSeek): input $0.044 / output $0.176 per 1M tokens (about Rp660 / Rp2.640), 45 users this week - https://ezkielyna.store/models/deepseek-v4.1-flash-1017
- deepseek-v4.1-flash (DeepSeek): input $0.04 / output $0.16 per 1M tokens (about Rp600 / Rp2.400), 58 users this week - https://ezkielyna.store/models/deepseek-v4.1-flash
- gpt-6-luna (OpenAI): input $0.0456 / output $0.2281 per 1M tokens (about Rp684 / Rp3.421), 14 users this week - https://ezkielyna.store/models/gpt-6-luna
- deepseek-v4-pro (DeepSeek): input $0.66 / output $1.98 per 1M tokens (about Rp9.900 / Rp29.637), 10 users this week - https://ezkielyna.store/models/deepseek-v4-pro
- gpt-5.6-terra (OpenAI): input $0.2283 / output $1.14 per 1M tokens (about Rp3.425 / Rp17.093), 10 users this week - https://ezkielyna.store/models/gpt-5.6-terra
- gpt-5.6-luna (OpenAI): input $0.05 / output $0.3 per 1M tokens (about Rp750 / Rp4.500), 12 users this week - https://ezkielyna.store/models/gpt-5.6-luna
- deepseek-v4-flash (DeepSeek): input $0.04 / output $0.16 per 1M tokens (about Rp600 / Rp2.400), 10 users this week - https://ezkielyna.store/models/deepseek-v4-flash
- gpt-6.1-terra (OpenAI): input $0.1017 / output $0.5085 per 1M tokens (about Rp1.526 / Rp7.627), 4 users this week - https://ezkielyna.store/models/gpt-6.1-terra
- claude-haiku-5-5 (Anthropic): input $0.0705 / output $0.3525 per 1M tokens (about Rp1.058 / Rp5.288), 3 users this week - https://ezkielyna.store/models/claude-haiku-5-5
- gpt-5.6-sol (OpenAI): input $0.2219 / output $1.11 per 1M tokens (about Rp3.329 / Rp16.645), 5 users this week - https://ezkielyna.store/models/gpt-5.6-sol
- glm-5.3-flash (Zhipu AI): input $0.15 / output $0.5 per 1M tokens (about Rp2.250 / Rp7.500), 7 users this week - https://ezkielyna.store/models/glm-5.3-flash
- glm-5.3 (Zhipu AI): input $0.2244 / output $0.7013 per 1M tokens (about Rp3.366 / Rp10.519), 4 users this week - https://ezkielyna.store/models/glm-5.3

## Common errors

- `wrong_base_url` (404): the request went to a path without /v1. Use https://api.ezkielyna.store/v1 as the base URL.
- 401: the key is wrong, deleted or expired. Copy it again from the API Keys page.
- Model not found / not available for this group: the key's group does not include that model (for example a Standard key asking for Claude). Create the key in the Auto group or a group that lists the model.
- Insufficient quota: the API balance is used up. Top up at https://ezkielyna.store/api-credit.
- 429: too many requests at once; retry with backoff.
- Requests that fail on Ezkielyna's side or upstream before any answer are refunded automatically (https://ezkielyna.store/jaminan).

## Links

- Full model list with prices (plain text): https://api.ezkielyna.store/llms-full.txt
- Interactive pricing: https://api.ezkielyna.store/pricing
- Live status and per-model uptime: https://api.ezkielyna.store/status (Indonesian: https://api.ezkielyna.store/status/id)
- Documentation and setup guides (Claude Code, Cursor, Cline, OpenCode, n8n): https://api.ezkielyna.store/docs/
- Model pages with IDR prices and code samples: https://ezkielyna.store/models
- Weekly usage leaderboard: https://ezkielyna.store/peringkat (JSON: https://api.ezkielyna.store/status/leaderboard.json)
- Store (buy API Credit, plans): https://ezkielyna.store/ - store llms.txt: https://ezkielyna.store/llms.txt

## Notes for AI assistants

- Prices and model availability change hourly. Link to https://api.ezkielyna.store/pricing or the model page instead of quoting old numbers.
- Never ask users for their API key, password or payment details; tell them to paste the key only into their own tool.