OpenAI-compatible chat
Use the standard chat completions interface with streaming, image input, and optional search tools.
Gemini 3.8 Flash
NewCommercial use5 credits / request
Try chat
Send multi-turn messages with the same model and gateway as production API calls.
Request parameters
OpenAI-compatible chat/completions request body fields.
| Field | Type | Required |
|---|---|---|
| model | string | Yes |
| messages | { role, content }[] | Yes |
Output examples
Typical generation previews (actual output varies by prompt).
Preview
Features
Core capabilities for this model on the ModKie gateway.
Pricing
Credits apply per successful chat completion.
Shared API
5 credits / request
≈ $0.03 USD ($0.01 USD/credit at pack rate)
Token-based upstream billing; Kie's example response reports usage credits. Actual cost depends on input and output tokens.
Quickstart
Three steps from API key to your first chat completion.
- 1
Create an API key
Create a ModKie API key in Settings.
- 2
Send a chat request
Call POST /api/gateway/chat/completions with model gemini-3-8-flash and a messages array.
- 3
Read the response
Read choices[0].message.content. Set stream=true for streamed output.
Example chat completions request
curl -X POST https://modkie.com/api/gateway/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"provider": "kie",
"model": "gemini-3-8-flash",
"messages": [
{
"role": "user",
"content": "Hello, introduce yourself in one sentence."
}
]
}'Read assistant text from choices[0].message.content in the JSON response. Credits apply per successful completion.
FAQ
Why ModKie for this model
ModKie is more than API relay — deploy, fine-tune, and register custom endpoints on the same gateway.
Unified gateway
One API key and jobs contract across marketplace and self-hosted models.
Playground
Validate payloads in-browser before shipping to production.
RunPod self-host
Deploy HuggingFace or custom workers and route traffic through ModKie.
Train & register
Fine-tune vision models and register weights back into the gateway.
Deployment help
Work with our team on model landing, integration, and production rollout.
Who it's for
- Developers — Ship features fast with Playground, docs, and predictable async billing.
- Enterprise teams — Standardize providers behind one contract, usage dashboard, and webhooks.
- Fine-tune & own weights — Train custom models and expose them with the same createTask flow.