Cosmos 3 Super API - AI Image & Video Generation APIs: Pricing, Documentation

Cosmos 3 Super API delivers a unified, high-performance integration layer designed to streamline complex data workflows and accelerate digital transformation across enterprise ecosystems. This robust solution offers developers seamless connectivity, exceptional uptime, and low-latency processing to handle demanding workloads without compromising on speed or reliability. With its intuitive architecture and advanced security protocols, the platform simplifies API management, automates routine tasks, and scales effortlessly to meet growing business demands. By eliminating operational bottlenecks and ensuring data consistency across various internal endpoints, it empowers organizations to optimize their software architecture and unlock maximum technical efficiency.

Get API Key

Models Version

WELCOME BONUS

Get $5 Free Credit on First Payment

No strings attached — add funds and get $5 bonus instantly

Claim Your $5 →

Cosmos 3 Super API Documentation

https://gateway.pixazo.ai/cosmos-3-super-text-to-image/v1/cosmos-3-super-text-to-image-request

Authentication

All requests require an API key passed via header.

Header	Type	Required	Description
Ocp-Apim-Subscription-Key	string	Yes	Your API subscription key

Cosmos 3 Super Text to Image generate request

Request Code

POST https://gateway.pixazo.ai/cosmos-3-super-text-to-image/v1/cosmos-3-super-text-to-image-request
Content-Type: application/json
Cache-Control: no-cache
Ocp-Apim-Subscription-Key: YOUR_API_KEY

{
  "prompt": "A futuristic mountain observatory under a star-filled sky, cinematic lighting, photorealistic",
  "negative_prompt": "",
  "image_size": "square_hd",
  "num_inference_steps": 28,
  "guidance_scale": 4,
  "num_images": 1,
  "output_format": "jpeg",
  "enable_prompt_expansion": false,
  "enable_safety_checker": true,
  "sync_mode": false
}

import requests

url = "https://gateway.pixazo.ai/cosmos-3-super-text-to-image/v1/cosmos-3-super-text-to-image-request"
headers = {
    "Content-Type": "application/json",
    "Cache-Control": "no-cache",
    "Ocp-Apim-Subscription-Key": "YOUR_API_KEY"
}
data = {
    "prompt": "A futuristic mountain observatory under a star-filled sky, cinematic lighting, photorealistic",
    "negative_prompt": "",
    "image_size": "square_hd",
    "num_inference_steps": 28,
    "guidance_scale": 4,
    "num_images": 1,
    "output_format": "jpeg",
    "enable_prompt_expansion": false,
    "enable_safety_checker": true,
    "sync_mode": false
}

response = requests.post(url, json=data, headers=headers)
print(response.json())

const url = "https://gateway.pixazo.ai/cosmos-3-super-text-to-image/v1/cosmos-3-super-text-to-image-request";
const headers = {
  "Content-Type": "application/json",
  "Cache-Control": "no-cache",
  "Ocp-Apim-Subscription-Key": "YOUR_API_KEY"
};
const data = {
  "prompt": "A futuristic mountain observatory under a star-filled sky, cinematic lighting, photorealistic",
  "negative_prompt": "",
  "image_size": "square_hd",
  "num_inference_steps": 28,
  "guidance_scale": 4,
  "num_images": 1,
  "output_format": "jpeg",
  "enable_prompt_expansion": false,
  "enable_safety_checker": true,
  "sync_mode": false
};

fetch(url, {
  method: "POST",
  headers: headers,
  body: JSON.stringify(data)
})
.then(response => response.json())
.then(data => console.log(data));

curl -X POST "https://gateway.pixazo.ai/cosmos-3-super-text-to-image/v1/cosmos-3-super-text-to-image-request" \
  -H "Content-Type: application/json" \
  -H "Cache-Control: no-cache" \
  -H "Ocp-Apim-Subscription-Key: YOUR_API_KEY" \
  --data-raw '{
    "prompt": "A futuristic mountain observatory under a star-filled sky, cinematic lighting, photorealistic",
    "negative_prompt": "",
    "image_size": "square_hd",
    "num_inference_steps": 28,
    "guidance_scale": 4,
    "num_images": 1,
    "output_format": "jpeg",
    "enable_prompt_expansion": false,
    "enable_safety_checker": true,
    "sync_mode": false
  }'

Output

{
  "request_id": "cosmos-3-super-text-to-image_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "QUEUED",
  "polling_url": "https://gateway.pixazo.ai/v2/requests/status/cosmos-3-super-text-to-image_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
}

Try Now

Webhook (Optional)

Add the X-Webhook-URL header to your generate request to receive a POST callback instead of polling.

X-Webhook-URL: https://your-server.com/webhook/callback

Request Parameters - Cosmos 3 Super Text to Image generate request

Parameter	Required	Type	Default	Allowed values / range	Description
prompt	Yes	string	—	1–4096 characters	Text prompt for image generation (1-4096 characters).
negative_prompt	No	string	""	≤ 1024 characters	Content to steer away from during generation (max 1024 characters).
enable_prompt_expansion	No	boolean	false	—	Rewrites your prompt into the dense, structured format the model was trained on before generation, which usually improves detail and prompt adherence. If expansion fails, your raw prompt is used.
enable_agentic_generation	No	boolean	false	—	Enable iterative generation and comparison of multiple candidate images. Each candidate is billed separately.
agentic_max_iterations	No	integer	2	1–3	Maximum refinement rounds (1-3). Only applies if enable_agentic_generation is true.
agentic_samples_per_iteration	No	integer	2	1–3	Number of candidate images generated per iteration (1-3). Only applies if enable_agentic_generation is true.
agentic_early_stop	No	boolean	true	—	Stop refinement early if a candidate is deemed a strong match. Only applies if enable_agentic_generation is true.
image_size	No	string or object	"square_hd"	"square_hd", "square", "portrait_4_3", "portrait_16_9", "landscape_4_3", "landscape_16_9", or {width: integer, height: integer} with each edge 512–1280 px in multiples of 16	Size of the generated image. Use a preset (square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9) or a custom object such as {"width": 1024, "height": 1024}. Each edge is clamped to 512-1280 px and snapped to a multiple of 16.
num_inference_steps	No	integer	28	1–50	Number of refinement steps the model runs while generating. Higher values refine detail and quality but increase processing time; lower values are faster.
guidance_scale	No	float	4	0–20	Controls how closely the output follows your prompt (prompt adherence). Higher values stick more strictly to the prompt; lower values allow more creative variation.
num_images	No	integer	1	1–4	Number of images to generate (1-4). Each generated image is billed separately.
output_format	No	string	"jpeg"	"jpeg", "png"	Output image format. Allowed: jpeg, png.
enable_safety_checker	No	boolean	true	—	Turns on content moderation for the input prompt and the generated images. Leave enabled unless you have a specific reason to disable it.
seed	No	integer	—	—	Random seed that controls the generation's randomness. Reuse the same seed with identical settings to reproduce the same result; leave it empty for a different output each time.
sync_mode	No	boolean	false	—	If on, the finished file is sent back directly instead of as a download link. Handy for quick tests; for normal use, leave it off to get a download link.

Example Request

{
  "prompt": "A futuristic mountain observatory under a star-filled sky, cinematic lighting, photorealistic",
  "negative_prompt": "",
  "image_size": "square_hd",
  "num_inference_steps": 28,
  "guidance_scale": 4,
  "num_images": 1,
  "output_format": "jpeg",
  "enable_prompt_expansion": false,
  "enable_safety_checker": true,
  "sync_mode": false
}

Response

{
  "request_id": "cosmos-3-super-text-to-image_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "QUEUED",
  "polling_url": "https://gateway.pixazo.ai/v2/requests/status/cosmos-3-super-text-to-image_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
}

Request Headers

Header	Value
Content-Type	application/json
Cache-Control	no-cache
Ocp-Apim-Subscription-Key	YOUR_API_KEY

Response Handling

Common status codes.

Code	Meaning
202	Accepted — Request queued
400	Bad Request
401	Unauthorized
402	Insufficient Balance
403	Forbidden
429	Too Many Requests
500	Internal Server Error

Error Responses

Queue system errors and model validation errors.

Queue System Errors

// 402 — Insufficient balance
{
  "error": "Insufficient Balance",
  "message": "Your wallet does not have enough balance."
}

// 400 — Model not found
{
  "error": "Model not found",
  "message": "Model 'cosmos-3-super-text-to-image' not found or is disabled"
}

Error via Status/Webhook

{
  "request_id": "cosmos-3-super-text-to-image_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "cosmos-3-super-text-to-image",
  "error": "Description of the error",
  "output": null
}

Retrieving Results

Poll the universal status endpoint to check progress and retrieve results.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_API_KEY

cURL Example

curl -H "Ocp-Apim-Subscription-Key: YOUR_API_KEY" \
  "https://gateway.pixazo.ai/v2/requests/status/cosmos-3-super-text-to-image_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"

Response (Completed)

{
  "request_id": "cosmos-3-super-text-to-image_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "cosmos-3-super-text-to-image",
  "error": null,
  "output": {
    "media_url": ["https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/cosmos-3-super-text-to-image_019dxxxx/output.jpg"],
    "media_type": "image/jpeg"
  },
  "created_at": "2026-03-31T10:00:00.000Z",
  "updated_at": "2026-03-31T10:00:15.000Z",
  "completed_at": "2026-03-31T10:00:15.000Z"
}

Response Fields

Field	Type	Description
request_id	string	Unique request identifier
status	string	QUEUED, PROCESSING, COMPLETED, FAILED, or ERROR
model_id	string	Model that processed the request
error	string\|null	Error message if failed
output.media_url	array	URLs to generated media (R2 CDN)
output.media_type	string	MIME type of the output
created_at	string	When request was created
completed_at	string	When request completed
polling_url	string	Status URL (initial response only)

Status Values

Status	Description
QUEUED	Request accepted, waiting to be processed
PROCESSING	Being processed by the model
COMPLETED	Done — output contains the result
FAILED	Failed — check error field
ERROR	System error — not charged

Status Flow

QUEUED → PROCESSING → COMPLETED
                    → FAILED
                    → ERROR

Typical Workflow

Send a generate request to the API endpoint
Save the request_id from the response
Poll every 5-10 seconds: GET /v2/requests/status/{request_id}
When status is "COMPLETED", download from output.media_url

Tip: Use X-Webhook-URL header to get a callback instead of polling.

Cosmos 3 Super API Pricing

Usage	Price (USD)
Per image	$0.04
Prompt expansion (per request)	$0.02

Image to Video

Cosmos 3 Super API Documentation

https://gateway.pixazo.ai/cosmos-3-super-image-to-video/v1/cosmos-3-super-image-to-video-request

Authentication

All requests require an API key passed via header.

Header	Type	Required	Description
Ocp-Apim-Subscription-Key	string	Yes	Your API subscription key

Cosmos 3 Super Image to Video generate request

Request Code

POST https://gateway.pixazo.ai/cosmos-3-super-image-to-video/v1/cosmos-3-super-image-to-video-request
Content-Type: application/json
Cache-Control: no-cache
Ocp-Apim-Subscription-Key: YOUR_API_KEY

{
  "prompt": "Cinematic slow camera push-in; the subject animates naturally",
  "image_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/Image.jpeg",
  "num_frames": 189,
  "frames_per_second": 24,
  "num_inference_steps": 28,
  "guidance_scale": 6,
  "enable_prompt_expansion": true,
  "enable_safety_checker": true,
  "sync_mode": false
}

import requests

url = "https://gateway.pixazo.ai/cosmos-3-super-image-to-video/v1/cosmos-3-super-image-to-video-request"
headers = {
    "Content-Type": "application/json",
    "Cache-Control": "no-cache",
    "Ocp-Apim-Subscription-Key": "YOUR_API_KEY"
}
data = {
    "prompt": "Cinematic slow camera push-in; the subject animates naturally",
    "image_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/Image.jpeg",
    "num_frames": 189,
    "frames_per_second": 24,
    "num_inference_steps": 28,
    "guidance_scale": 6,
    "enable_prompt_expansion": true,
    "enable_safety_checker": true,
    "sync_mode": false
}

response = requests.post(url, json=data, headers=headers)
print(response.json())

const url = "https://gateway.pixazo.ai/cosmos-3-super-image-to-video/v1/cosmos-3-super-image-to-video-request";
const headers = {
  "Content-Type": "application/json",
  "Cache-Control": "no-cache",
  "Ocp-Apim-Subscription-Key": "YOUR_API_KEY"
};
const data = {
  "prompt": "Cinematic slow camera push-in; the subject animates naturally",
  "image_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/Image.jpeg",
  "num_frames": 189,
  "frames_per_second": 24,
  "num_inference_steps": 28,
  "guidance_scale": 6,
  "enable_prompt_expansion": true,
  "enable_safety_checker": true,
  "sync_mode": false
};

fetch(url, {
  method: "POST",
  headers: headers,
  body: JSON.stringify(data)
})
.then(response => response.json())
.then(data => console.log(data));

curl -X POST "https://gateway.pixazo.ai/cosmos-3-super-image-to-video/v1/cosmos-3-super-image-to-video-request" \
  -H "Content-Type: application/json" \
  -H "Cache-Control: no-cache" \
  -H "Ocp-Apim-Subscription-Key: YOUR_API_KEY" \
  --data-raw '{
    "prompt": "Cinematic slow camera push-in; the subject animates naturally",
    "image_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/Image.jpeg",
    "num_frames": 189,
    "frames_per_second": 24,
    "num_inference_steps": 28,
    "guidance_scale": 6,
    "enable_prompt_expansion": true,
    "enable_safety_checker": true,
    "sync_mode": false
  }'

Output

{
  "request_id": "cosmos-3-super-image-to-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "QUEUED",
  "polling_url": "https://gateway.pixazo.ai/v2/requests/status/cosmos-3-super-image-to-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
}

Try Now

Webhook (Optional)

Add the X-Webhook-URL header to your generate request to receive a POST callback instead of polling.

X-Webhook-URL: https://your-server.com/webhook/callback

Request Parameters - Cosmos 3 Super Image to Video generate request

Parameter	Required	Type	Default	Allowed values / range	Description
prompt	Yes	string	—	non-empty; ≤ 4096 chars	Text describing motion and scene (max 4096 characters). Must describe dynamic elements such as camera movement, object motion, or environmental changes.
image_url	Yes	string	—	—	URL of the conditioning first-frame image. Must be publicly accessible.
negative_prompt	No	string	NVIDIA's recommended i2v negative prompt (long; see description)	≤ 2048 chars	Content to steer the generation away from (artifacts, unwanted motion). If you omit it, NVIDIA's long recommended image-to-video negative prompt (covering compression artifacts, noise, shaky footage, colour banding, unnatural motion, etc.) is applied automatically. Pass an empty string ("") to disable it, or your own text to override it. Max 2048 characters.
enable_prompt_expansion	No	boolean	true	—	Rewrites your prompt into the dense caption the model was trained on, using a reasoner that also looks at the first frame. Usually improves motion and prompt adherence; if expansion fails, your raw prompt is used.
enable_agentic_generation	No	boolean	false	—	Enables an iterative refinement loop (prompt upsampling, candidate videos, automatic critique, prompt rewrite). Each candidate is a full render, so this is substantially slower and costlier than a single generation.
agentic_max_iterations	No	integer	2	1–3	Maximum number of agentic refinement stages (1-3). Only applies if enable_agentic_generation is true.
agentic_samples_per_iteration	No	integer	2	1–3	Number of candidate videos generated and judged per agentic iteration (1-3); the best one advances to the next round. Only applies if enable_agentic_generation is true.
agentic_early_stop	No	boolean	true	—	Stop the agentic loop early once a candidate clears the quality threshold. Only applies if enable_agentic_generation is true.
image_size	No	string or object	{"width":832,"height":480}	"square_hd", "square", "portrait_4_3", "portrait_16_9", "landscape_4_3", "landscape_16_9", or {width: integer, height: integer}	Size of the generated video. Use a preset (square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9) or a custom object such as {"width": 832, "height": 480}. The request is clamped and snapped to the nearest supported tier (256p / 480p / 720p) and aspect ratio.
num_frames	No	integer	189	5–189	Number of frames to generate. More frames yield a longer video: duration ≈ num_frames ÷ frames_per_second (189 frames at 24 fps ≈ 7.9 seconds).
frames_per_second	No	integer	24	4–60	Frame rate of the output video. Together with num_frames this sets the video duration.
num_inference_steps	No	integer	28	1–50	Number of refinement steps the model runs while generating. Higher values refine detail and quality but increase processing time; lower values are faster.
guidance_scale	No	float	6	0–20	Controls how closely the output follows your prompt (prompt adherence). Higher values stick more strictly to the prompt; lower values allow more creative variation.
enable_safety_checker	No	boolean	true	—	Turns on content moderation for the input prompt and the input image. Leave enabled unless you have a specific reason to disable it.
seed	No	integer	—	—	Random seed that controls the generation's randomness. Reuse the same seed with identical settings to reproduce the same result; leave it empty for a different output each time.
sync_mode	No	boolean	false	—	If on, the finished file is sent back directly instead of as a download link. Handy for quick tests; for normal use, leave it off to get a download link.

Content Item Types & Limits

Type	Max	Format / Size	Description
image	1	JPG, PNG, WEBP	Input image to animate.

Example Request

{
  "prompt": "Cinematic slow camera push-in; the subject animates naturally",
  "image_url": "https://pub-582b7213209642b9b995c96c95a30381.r2.dev/Image.jpeg",
  "num_frames": 189,
  "frames_per_second": 24,
  "num_inference_steps": 28,
  "guidance_scale": 6,
  "enable_prompt_expansion": true,
  "enable_safety_checker": true,
  "sync_mode": false
}

Response

{
  "request_id": "cosmos-3-super-image-to-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "QUEUED",
  "polling_url": "https://gateway.pixazo.ai/v2/requests/status/cosmos-3-super-image-to-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
}

Request Headers

Header	Value
Content-Type	application/json
Cache-Control	no-cache
Ocp-Apim-Subscription-Key	YOUR_API_KEY

Response Handling

Common status codes.

Code	Meaning
202	Accepted — Request queued
400	Bad Request
401	Unauthorized
402	Insufficient Balance
403	Forbidden
429	Too Many Requests
500	Internal Server Error

Error Responses

Queue system errors and model validation errors.

Queue System Errors

// 402 — Insufficient balance
{
  "error": "Insufficient Balance",
  "message": "Your wallet does not have enough balance."
}

// 400 — Model not found
{
  "error": "Model not found",
  "message": "Model 'cosmos-3-super-image-to-video' not found or is disabled"
}

Error via Status/Webhook

{
  "request_id": "cosmos-3-super-image-to-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "ERROR",
  "model_id": "cosmos-3-super-image-to-video",
  "error": "Description of the error",
  "output": null
}

Retrieving Results

Poll the universal status endpoint to check progress and retrieve results.

Endpoint

GET https://gateway.pixazo.ai/v2/requests/status/{request_id}
Ocp-Apim-Subscription-Key: YOUR_API_KEY

cURL Example

curl -H "Ocp-Apim-Subscription-Key: YOUR_API_KEY" \
  "https://gateway.pixazo.ai/v2/requests/status/cosmos-3-super-image-to-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"

Response (Completed)

{
  "request_id": "cosmos-3-super-image-to-video_019dxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "status": "COMPLETED",
  "model_id": "cosmos-3-super-image-to-video",
  "error": null,
  "output": {
    "media_url": ["https://pub-582b7213209642b9b995c96c95a30381.r2.dev/v1/cosmos-3-super-image-to-video_019dxxxx/output.mp4"],
    "media_type": "video/mp4"
  },
  "created_at": "2026-03-31T10:00:00.000Z",
  "updated_at": "2026-03-31T10:00:15.000Z",
  "completed_at": "2026-03-31T10:00:15.000Z"
}

Response Fields

Field	Type	Description
request_id	string	Unique request identifier
status	string	QUEUED, PROCESSING, COMPLETED, FAILED, or ERROR
model_id	string	Model that processed the request
error	string\|null	Error message if failed
output.media_url	array	URLs to generated media (R2 CDN)
output.media_type	string	MIME type of the output
created_at	string	When request was created
completed_at	string	When request completed
polling_url	string	Status URL (initial response only)

Status Values

Status	Description
QUEUED	Request accepted, waiting to be processed
PROCESSING	Being processed by the model
COMPLETED	Done — output contains the result
FAILED	Failed — check error field
ERROR	System error — not charged

Status Flow

QUEUED → PROCESSING → COMPLETED
                    → FAILED
                    → ERROR

Typical Workflow

Send a generate request to the API endpoint
Save the request_id from the response
Poll every 5-10 seconds: GET /v2/requests/status/{request_id}
When status is "COMPLETED", download from output.media_url

Tip: Use X-Webhook-URL header to get a callback instead of polling.

Cosmos 3 Super API Pricing

Your request will cost $0.05 per second.

⚡ Performance

Live usage measured on Pixazo's gateway, split by model version. Generation time is how long a generation takes end-to-end (lower is better). Success rate is the percent of generations that complete (higher is better).

Show data for the last

Generations

1,700last 30d

~57 per day

Success rate

100.0%

of completed generations

Generation time

5.4minavg

p95 6.6min

Requests

Jun 27max 400Jul 26

Cosmos 3 SuperAvg 57/day

Generation Time

Jun 27max 12.6minJul 26

Cosmos 3 SuperAvg 5.4min

Error Rate

Jun 27max 5.0%Jul 26

Cosmos 3 SuperAvg 0.0%

〰 Uptime

Percent of generations that succeeded over the selected period, per model version.

Avg. Success Rate (30d)

100.00%

across all generations of this model family

Uptime

Jun 27max 100%Jul 26

SuccessfulAvg 100.00%