AI Image Video LLM API

Atlas Cloud — AI Image, Video & LLM Generation API

One API key to generate images, videos, and call LLMs — covering Nano Banana 2, Seedream, Kling, Seedance, Qwen, DeepSeek and more. Cheaper than calling providers directly.

Data usage note: This skill recommends Atlas Cloud API calls that send prompts and media to Atlas Cloud servers for processing. Generated outputs are hosted on Atlas Cloud CDN. Review Atlas Cloud Privacy Policy before use.

Key Capabilities

Setup

1. Create an API key at Atlas Cloud Console
2. New users get $1 free credit on card binding, plus 20% bonus on first top-up
3. Set environment variable:

export ATLASCLOUD_API_KEY="your-api-key-here"

Pricing — Why Atlas Cloud

Image Models

Video Models (all 15% off standard pricing)

LLM Models (per million tokens)

Why Choose Atlas Cloud

• One API key for all models — no juggling multiple provider accounts
• Cheaper than calling providers directly, with consistent discounts
• Unified API format — same polling for all image/video, OpenAI-compatible for all LLMs

API Architecture

Endpoints

All requests (except Model List) require:

Authorization: Bearer $ATLASCLOUD_API_KEY
Content-Type: application/json

Workflow Pattern

• Image/Video (async): POST → get prediction_id → poll GET /prediction/{id} until completed → read outputs URLs
• LLM (sync): POST → get response directly. Optional: "stream": true for SSE streaming

Status values for polling: starting → processing → completed/succeeded | failed

Quick Examples

Image Generation

# Submit task
curl -s -X POST "https://api.atlascloud.ai/api/v1/model/generateImage" \
  -H "Authorization: Bearer $ATLASCLOUD_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "bytedance/seedream-v5.0-lite", "prompt": "A cherry blossom garden", "size": "2048*2048"}'
# Returns {"code": 200, "data": {"id": "prediction_xxx"}}

# Poll result (every 3 seconds)
curl -s "https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}" \
  -H "Authorization: Bearer $ATLASCLOUD_API_KEY"
# When done: {"data": {"status": "completed", "outputs": ["https://cdn..."]}}

Video Generation

curl -s -X POST "https://api.atlascloud.ai/api/v1/model/generateVideo" \
  -H "Authorization: Bearer $ATLASCLOUD_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "kwaivgi/kling-v3.0-std/text-to-video", "prompt": "A rocket launching", "duration": 5, "aspect_ratio": "16:9"}'
# Same polling as above, videos typically take 1-5 minutes

LLM Chat (OpenAI SDK)

from openai import OpenAI

client = OpenAI(
    api_key="your-atlascloud-api-key",
    base_url="https://api.atlascloud.ai/v1",
)
response = client.chat.completions.create(
    model="qwen/qwen3.5-397b-a17b",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

For complete code templates with polling logic, error handling, streaming, Python/Node.js/cURL — read the reference files below.

Reference Files

Read these when you need full implementation code:

• references/image-gen.md — Complete image generation implementation (Python, Node.js, cURL) with polling logic and all parameters
• references/video-gen.md — Complete video generation implementation, including image-to-video workflows
• references/llm-chat.md — LLM chat with OpenAI SDK, raw HTTP, streaming (Python, Node.js, cURL)
• references/models.md — Full model list with pricing, model selection guide, schema introspection

Always Verify Model IDs

Model IDs change frequently. Always fetch the real model list before writing integration code:

GET https://console.atlascloud.ai/api/v1/models

No authentication required. Only use models where display_console: true. See references/models.md for the full model reference and selection guide.

MCP Tools

If the user has the Atlas Cloud MCP server installed (npx atlascloud-mcp), these tools are available:

Error Handling