API Documentation

overview

The GoonsAI API provides programmatic access to GPU compute resources for multimedia generation. Integrate text, image, video, and audio generation into your applications with predictable, usage-based pricing.

Base URL: https://api.goonsai.com/v1

authentication

All requests require an API key passed in the header:

Authorization: Bearer YOUR_API_KEY

API keys are issued upon approval. Keep your key secure — it provides access to your compute allocation.

endpoints

text generation

POST /generate/text
{
  "model": "goonsai-text-v1",
  "prompt": "your prompt",
  "max_tokens": 2048,
  "temperature": 0.7
}

image generation

POST /generate/image
{
  "model": "goonsai-image-v1",
  "prompt": "your prompt",
  "width": 1024,
  "height": 1024,
  "steps": 30
}

video generation

POST /generate/video
{
  "model": "goonsai-video-v1",
  "prompt": "your prompt",
  "duration": 5,
  "fps": 24,
  "resolution": "720p"
}

image-to-video

POST /generate/video/from-image
{
  "model": "goonsai-i2v-v1",
  "image": "base64_encoded_image",
  "prompt": "motion description",
  "duration": 5
}

audio generation

POST /generate/audio
{
  "model": "goonsai-audio-v1",
  "prompt": "your prompt",
  "duration": 30,
  "format": "mp3"
}

available models

Model	Type	Minutes/Request
`goonsai-text-v1`	Text	~0.01
`goonsai-image-v1`	Image	~0.05
`goonsai-image-hd-v1`	Image (high-res)	~0.12
`goonsai-video-v1`	Video	~0.8
`goonsai-video-hd-v1`	Video (1080p)	~1.5
`goonsai-i2v-v1`	Image-to-Video	~1.0
`goonsai-audio-v1`	Audio	~0.15

minutes & cost calculation

Usage is measured in compute minutes — the actual GPU time consumed by your request. This provides predictable, transparent pricing.

calculation

cost = base_minutes × quantization_multiplier × resolution_multiplier

Example: A 5-second video at 720p with standard quantization:

0.8 base × 1.0 quant × 1.0 resolution = 0.8 minutes

checking balance

GET /account/balance

Response:
{
  "minutes_remaining": 142.5,
  "minutes_used_today": 12.3
}

usage logs

GET /account/usage?from=2026-01-01&to=2026-01-31

Response:
{
  "total_minutes": 847.2,
  "requests": 1203,
  "by_model": { ... }
}

quantization levels

Reduce compute costs by selecting lower quantization precision. Trade-off between quality and cost.

Level	Precision	Cost Multiplier	Use Case
`fp16`	16-bit	1.0×	Maximum quality
`int8`	8-bit	0.7×	Balanced
`int4`	4-bit	0.5×	Drafts, previews

POST /generate/image
{
  "model": "goonsai-image-v1",
  "prompt": "your prompt",
  "quantization": "int8"
}

batch processing

Submit multiple requests in a single call for efficient processing.

POST /generate/batch
{
  "requests": [
    { "model": "goonsai-image-v1", "prompt": "prompt 1" },
    { "model": "goonsai-image-v1", "prompt": "prompt 2" },
    { "model": "goonsai-image-v1", "prompt": "prompt 3" }
  ],
  "webhook": "https://your-server.com/callback"
}

Response:
{
  "batch_id": "batch_abc123",
  "status": "queued",
  "estimated_minutes": 0.15
}

Results are delivered to your webhook or retrieved via:

GET /batch/{batch_id}

content moderation

Configure guardrails and content filtering for your application.

guardrail configuration

POST /config/guardrails
{
  "block_adult": true,
  "block_violence": true,
  "block_hate": true,
  "custom_blocklist": ["term1", "term2"]
}

prompt filtering

POST /config/filters
{
  "pre_filter": true,
  "sensitivity": "strict",
  "reject_on_match": true
}

vision safety

For image-to-video and image inputs, configure vision analysis:

POST /config/vision
{
  "scan_inputs": true,
  "block_prohibited": true,
  "log_rejections": true
}

Guardrail configurations are applied at the account level. Contact support for enterprise-level customization.

prompt adherence

Control how closely the model follows your prompt versus creative interpretation.

POST /generate/image
{
  "model": "goonsai-image-v1",
  "prompt": "your prompt",
  "adherence": 0.85,
  "cfg_scale": 7.5
}

Parameter	Range	Effect
`adherence`	0.0 - 1.0	Prompt following strength
`cfg_scale`	1.0 - 20.0	Classifier-free guidance

sandbox environment

Test your integration without consuming production minutes.

Sandbox URL: https://sandbox.api.goonsai.com/v1

limitations

Lower resolution outputs (max 512px for images)
Watermarked results
Rate limited to 10 requests/minute
No batch processing

sandbox credentials

Use your production API key with the sandbox URL. Sandbox usage is logged separately and does not affect your balance.

rate limits

Tier	Requests/min	Concurrent
Standard	60	5
Pro	300	20
Enterprise	Custom	Custom

Rate limit headers are included in all responses:

X-RateLimit-Limit: 60
X-RateLimit-Remaining: 45
X-RateLimit-Reset: 1704067200

error codes

Code	Meaning
`400`	Invalid request parameters
`401`	Invalid or missing API key
`402`	Insufficient minutes balance
`403`	Content blocked by guardrails
`429`	Rate limit exceeded
`503`	Service temporarily unavailable

get api access

API access is currently available by application. Submit your email and we'll be in touch with onboarding details.

support

For technical questions or integration help: [email protected]

api documentation