don't make us popular
Quack API
welcome here are the endpoints and how to use them

v0: ChatGPT 3.5 - 4 (randomly chooses based on demand) UNLIMITED

POST /v0/chat/completions Response:
{ "reply": "..." }
Limits: NONE unlimited requests per minute (yes probably the best choice for 99% of projects fast, unlimited use only problem is stablility)

v1: Chat/Completions (Text/Multimodal)

POST /v1/chat/completions Response:
{ "content": "...", "ms": 0  }
Allowed models: gpt-4o, gpt-4o-mini, gpt-4.1-nano, gpt-4.1-mini, gpt-4.1
Usage: Standard chat API suitable for general QA, reasoning, conversation, and multimodal (text/image) tasks. Cost-efficient or powerful depending on the model chosen.
Limits: 10 requests per minute.

v1: Text to Speech (Audio API)

POST /v1/audio/speech Response:
{ "res": "BASE64CODE", "ms": 0 }
Allowed models: gpt-4o-mini-tts-2025-12-15, tts-1
Usage: Converts text strings to spoken audio output optimized for speed.
Limits: 10 requests per minute.

v1: Speech to Text (Transcriptions)

POST /v1/audio/transcriptions
No file selected
Response:
{ "text": "...", "ms": 0 }
Allowed models: gpt-4o-mini-transcribe-2025-12-15, whisper-1
Usage: General speech recognition to convert audio input to text transcriptions. Note: Send audio as multipart/form-data.
Limits: 10 requests per minute.

v1: Image Edits

POST /v1/images/edits
No image selected
Response:
{ "created": 1589478378, "data": [ { "url": "..." } ], "ms": 0 }
Allowed models: gpt-image-1
Usage: Allows uploading a single image file under the image form property to perform edits.
Limits: 1 request per minute.

v2: Image Generation

POST /v2/images/generations Response:
{ "created": timestamp, "data": [ { "url": "..." } ], "ms": 0  }
Allowed models: gpt-image-1-mini
Usage: Cost-efficient image generation model from text prompts.
Limits: 1 request per minute.

v3: Base64 encoded Fast Image Generation

POST /v3/images/generations Response:
{ "output": "BASE64CODE", "ms": 0  }
Limits: 15 requests per minute (doesn't have to decode the base64 code so it's way faster than v2)

v4: Fast Audio Generation (AnyVoiceLab)

POST /v4/audio/generations Response:
{ "base64": "...", "ms": 0  }
Options: You can switch the gender of the voice by sending "gender": "male" or "female" in the request body.
Limits: 60 requests per minute.

v5: AI Text Humanizer UNLIMITED

POST /v5/chat/completions Response:
{ "0123...": "...", "ms": 0  }
Limits: 20 requests per minute

v6: AI Detector (gptcheck)

POST /v6/chat/completions Response:
{ "text": "...", "ms": 0  }
Limits: 5 requests per minute (good luck figuring out what the response even means i barely even understand the prediction logic it uses ¯_(ツ)_/¯)

v7: Translate (google) UNLIMITED :)

POST /v7/translate/ Response:
{ "text": "...", "ms": 0 }
Limits: NONE unlimited requests per minute :)

v8: Reliability AI

POST /v8/chat/completions Response:
{ "reply": "...", "ms": 0 }
Limits: 10 requests per minute self-healing - automatically switches models if one fails

v9: AI Image Describer

POST /v9/image/describer
No file selected
Response:
{ "description": "...", "ms": 0 }
Limits: 15 requests per minute. Upload an image above and click Test! Also accepts base64 via JSON: { "imageBase64": "data:image/png;base64,..." }
Note: Excessive abuse of the program may result in a permanent ban.