Bring language, image, and video models into your product. Explore APIs, compare prices, and connect with a Scenepond API key.
1{
2 "model": "deepseek-flash",
3 "messages": [
4 {
5 "role": "user",
6 "content": "Hello, Scenepond."
7 }
8 ]
9}Model IDs, protocols and configured rates fetched from the current catalog.
Animate a reference image into a video, guided by a text prompt.
alibaba/wan-3.0-prime/image-to-videoChat and reasoning through the OpenAI-compatible chat API.
deepseek-flashGenerate images from a text prompt.
meta/muse-image/text-to-imageAnimate a reference image into a video, guided by a text prompt.
blackforestlabs/flux-3/image-to-video/draftEdit a reference image with a text instruction.
openai/gpt-image-2.5/flare/editChat and reasoning through the OpenAI-compatible chat API.
deepseek-v4-proStart with the task. Choose the model and endpoint that fit it.
Power conversations, agents, and workflows.
Image APIsProduct imagery, editorial assets, and visual concepts through a model-specific API.
Video APIsBuild clips into creative workflows. Follow the model page for task submission and results.
Concept artwork. Model availability depends on the live catalog.
Use the gateway URL, your API key, and an enabled model ID. OpenAI-compatible models fit into existing SDK workflows; other protocols are listed per model.
https://scenepond-api-production.up.railway.app/v1deepseek-flashBearer $SCENEPOND_API_KEYKeep the API key on your server. Match the model ID and protocol to the catalog before sending a request.
Read the documentation1curl "https://scenepond-api-production.up.railway.app/v1/chat/completions" \
2 -H "Authorization: Bearer $SCENEPOND_API_KEY" \
3 -H "Content-Type: application/json" \
4 --data-binary @- <<'JSON'
5{
6 "model": "deepseek-flash",
7 "messages": [
8 {
9 "role": "user",
10 "content": "Hello!"
11 }
12 ],
13 "stream": false
14}
15JSONRun this example in a shell with cURL installed. Set SCENEPOND_API_KEY before running.
Coding tools, personal agents, and automation. Each guide shows the base URL, API key, model selection, and first request.
Use Scenepond as the model provider for your Hermes agent.
Agent runtimeSetup guideConnect your personal assistant to an enabled model through a custom provider.
Personal assistantSetup guideAdd Scenepond as a custom model API for coding and repository tasks.
Coding agentSetup guideConnect your terminal coding agent through an OpenAI-compatible provider.
Terminal coding agentSetup guideUse your Scenepond API key for coding tasks in the Cline extension.
IDE coding agentSetup guideConfigure a custom base URL and model for supported Cursor chat features.
IDE chatSetup guideUse an enabled chat model in an agent or knowledge-based application.
Apps & knowledgeSetup guideConnect an API request to a workflow, then use the response in the next node.
Workflow automationSetup guideThese are integration guides. Supported features depend on the tool version and the model protocol.
Four steps from choosing a model to reviewing your first charges.
Compare model capabilities, supported endpoints, and configured rates.
Recharge through the payment methods enabled for your account.
Name your key, set a spending limit, and keep the secret on your server.
Make your API call and review charges in your usage records.
Check a model's configured rate, set a key limit, and review the charge after a request.
Token models list input and output rates separately. Image and video billing depends on the model, protocol, and request settings.
View pricingUse a separate API key for each app. Set its spending limit and expiry, and disable or revoke it from the console when needed.
Manage API keysRequests use your account balance. Usage records show the model, recorded charge, and time. Asynchronous task charges may be adjusted after completion.
View usage recordsShort guides on requests, errors, and prompts.
Choose cURL, Python, or JavaScript. Set the endpoint, authenticate, and inspect the response.
Read the guideCheck authentication, model access, balance, and rate-limit responses before retrying.
Read the guideWrite specific prompts for image and video tasks, then refine against your own requirements.
Read the guideAnswers about billing, API keys, models, and getting started.
Read the documentation