Models/OpenAI/OpenAI API

Live6 featured endpoints

OpenAI API — GPT Image, Sora, and GPT Models

The OpenAI API on Muapi brings image generation and editing, Sora video, and GPT reasoning models into one developer workflow with one Muapi API key.Choose a family for the job: GPT Image for image creation and edits, Sora for video generation, or GPT models for text and multimodal reasoning. Each endpoint has its own request schema and pay-per-use pricing.

Get OpenAI API Key Explore GPT Image OpenAI API examples and comparison on GitHub

GPT reasoning

Text and multimodal reasoning endpoints.2 featured endpoints

ReasoningLiveFeatured

GPT reasoning

GPT 6 Astra

Multimodal reasoning endpoint for text and image input.

Text + image input

$10.00/M in · $50.00/M out

Try Model

ReasoningNew

GPT reasoning

GPT 6.1 Terra

Efficient visual reasoning, coding assistance, and writing with text and image input, optional web search, and six reasoning levels. Use polling or SSE streaming.

Text + image input

$1.00/M in · $5.00/M out

Try Model

GPT Image 2

Image generation and editing. See the family hub for the complete endpoint lineup.1 featured endpoints

ImageLive

GPT Image 2

Text to image generation.

Up to 4K

See live model pricing

Try Model

GPT Image 2.5

Additional image generation and editing variants.1 featured endpoints

ImageNew

GPT Image 2.5

GPT Image 2.5 Sunburst

Image generation in the GPT Image 2.5 family.

1K–4K

$0.06–$0.16/image

Try Model

Sora 2

Text and image to video generation.1 featured endpoints

VideoNew

Sora 2

Text to video generation.

720p

$0.80/request

Try Model

Sora 2 Pro

Higher resolution Sora video workflows.1 featured endpoints

VideoNew

Sora 2 Pro

Higher resolution text to video generation.

Up to 1080p

$2.40/request

Try Model

OpenAI model families on Muapi

OpenAI endpoints on Muapi span three distinct workflows: GPT Image for image generation and editing, Sora for video, and GPT models for reasoning. Use the family links below for complete variant lists and model-specific parameters.

Muapi uses asynchronous generation for media: submit a request, keep its request_id, then poll the shared predictions result endpoint. Text endpoints have their own model schemas; consult each model API page for accepted fields.

What you can build

Generate and edit images

Use GPT Image family endpoints for prompt-driven image creation and reference-based editing.

Create video with Sora

Submit text or image guided prompts to Sora video endpoints and retrieve completed outputs.

Reason over text and images

Call GPT reasoning models through dedicated endpoints for structured, multimodal tasks.

OpenAI API endpoint overview

Family	Input	Output	Pricing model	Full lineup
GPT Image	Prompt and optional reference images	Generated or edited image	Per generation; see endpoint pricing	/gpt-image
Sora	Prompt and optional reference image	Generated video	Per request; tier dependent	/sora
GPT reasoning	Text and supported multimodal inputs	Text response	Token based where listed	/gpt-6-astra
GPT 6.1 Terra	Prompt, optional image and system prompt	Text via polling or SSE	Standard: $1.00/M input · $5.00/M output; $0.0002 minimum	/playground/gpt-6-1-terra

Prices and available parameters vary by endpoint. The linked family pages show the complete lineup and current model-specific details. Terra cached input is $0.05/M. Above 272K input tokens, input/cache rates double and output costs $7.50/M.

How to call the OpenAI API

Submit a model-specific JSON request with your Muapi API key, save the request ID, and poll the shared result endpoint for completed text or media outputs. GPT 6.1 Terra also supports SSE streaming. Check each family page for exact parameters and response details.

1. Submit an image generation request

curl -X POST https://api.muapi.ai/api/v1/gpt-image-2-text-to-image \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Editorial product photo of a ceramic coffee cup on a warm wood table","aspect_ratio":"1:1","resolution":"2k"}'

# Response: {"request_id":"REQUEST_ID"}

Send a prompt to a GPT Image endpoint.

2. Submit a Sora video request

curl -X POST https://api.muapi.ai/api/v1/openai-sora-2-pro-text-to-video \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A slow tracking shot through a sunlit greenhouse","aspect_ratio":"16:9","duration":8,"resolution":"1080p"}'

# Response: {"request_id":"REQUEST_ID"}

Call Sora 2 Pro with a prompt and generation settings.

3. Submit a GPT reasoning request

curl -X POST https://api.muapi.ai/api/v1/gpt-6-astra \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Compare two database migration strategies and explain the tradeoffs.","reasoning_effort":"medium"}'

# Response: {"request_id":"REQUEST_ID"}

Call GPT 6 Astra with a prompt and reasoning effort.

4. Poll for the result

curl https://api.muapi.ai/api/v1/predictions/REQUEST_ID/result \
  -H "x-api-key: YOUR_API_KEY"

# Poll until status is completed, then read the generated asset URL from outputs.

Use the request ID to retrieve the completed output.

Call GPT 6.1 Terra

curl -X POST https://api.muapi.ai/api/v1/gpt-6-1-terra \
  -H "x-api-key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"Draft a concise proposal for automating invoice review.","reasoning_effort":"medium"}'

# Response: {"request_id":"REQUEST_ID"}

Send a prompt with optional image_url, system_prompt, web_search_switch, and reasoning_effort (none through max; default medium). Poll the returned request ID, or use /api/v1/gpt-6-1-terra/stream for SSE.

OpenAI API FAQ

Which OpenAI models are available on Muapi?

The page highlights GPT Image, Sora, and GPT reasoning endpoints. Follow the family links for the complete live lineup.

How do I call an OpenAI model?

Create a Muapi API key, send a POST request to the endpoint with its documented JSON body, then poll /api/v1/predictions/{request_id}/result for completion.

Do all OpenAI models use the same request fields?

No. Image, video, and reasoning endpoints have different schemas. Use the linked family hub and model API reference for the exact fields.

How is usage priced?

Pricing depends on the endpoint and model. Review the live endpoint pricing before submitting a request.

Can I use GPT 6.1 Terra through Muapi?

Yes. Call /api/v1/gpt-6-1-terra for polling or append /stream for live text over SSE. Terra supports text and image input, optional web search, and reasoning levels none, low, medium, high, xhigh, and max (default medium). Standard pricing is $1.00/M input, $0.05/M cached input, and $5.00/M output, with a $0.0002 minimum. Above 272K input tokens, input/cache rates double and output costs $7.50/M.

GPT Image APIBrowse GPT Image generation and editing endpoints.GPT Image 2.5 APIExplore GPT Image 2.5 variants.GPT Image 1.5 APISee the earlier GPT Image family hub.Sora APIBrowse Sora video generation endpoints.GPT 6 Astra APIReview GPT 6 Astra reasoning model details.GPT 6.1 Terra APIVisual reasoning, coding, and writing with optional images, web search, and SSE streaming.

Build with OpenAI models

Use one Muapi API key to integrate image, video, and reasoning endpoints into your application.

Get an API Key Explore GPT Image