Token-first security
Scope access per token, toggle features, and rotate secrets instantly without shipping new keys to clients.
Private AI Infrastructure
Build with Kent Wynn AI—secure LLM endpoints, managed tokens, zero guesswork.
Orchestrate chat, completions, embeddings, and tool-calling flows with an OpenAI-compatible surface. Control quota, monitor usage, and rely on Kent Wynn's managed infrastructure for consistent low-latency responses.
Daily check-in bonuses
Log in every day to add +1,000 free tokens to your Kent Wynn account.
NewGitHub launch bonus
Register with GitHub for the first time and unlock +100,000 tokens instantly.
Platform highlights
Signature model · kentwynn/reasoningEmbeddings · kentwynn/embeddingTool calling · functions & JSON schema
curl -X POST \
https://api.kentwynn.com/v1/chat/completions \
-H "Content-Type: application/json" \
-H "X-AI-Token: YOUR_TOKEN" \
-d '{
"model": "kentwynn/reasoning",
"messages": [
{ "role": "system", "content": "You are a helpful assistant." },
{ "role": "user", "content": "Summarise the Kent Wynn AI platform." }
],
"stream": false
}'Quickstart
1. Create your API token
Sign in to the Kent Wynn console, generate an API token, and copy it securely—tokens are only revealed once.
2. Pick a model alias
Use `kentwynn/reasoning` for chat-style workloads or `kentwynn/embedding` for search pipelines.
3. Call the endpoint
Send a POST request to the `/v1/*` endpoint with your token in the `X-AI-Token` header. Responses mirror OpenAI schemas.
4. Monitor usage
Track quotas and daily burn in the console or automate checks with the admin REST endpoints.
LLM compatibility
Chat, completions, embeddings, and responses expose an OpenAI-compatible contract for effortless integration.
Daily check-in rewards
Add +1,000 tokens to your account every 24 hours with the console check-in bonus—perfect for ongoing experiments.
Private hosting
Fully managed inference from Kent Wynn infrastructure with a polished, reliable API surface and predictable performance.
Live demo endpoints
Each button calls the /demo namespace on api.kentwynn.com with the sample payloads shown. Responses come directly from the hosted engines.
GETGET /demo/v1/models
List available demo model aliases and engine IDs.
curl -X 'GET' \
'https://api.kentwynn.com/demo/v1/models' \
-H 'accept: application/json'Core REST endpoints
All endpoints accept X-AI-Token and return JSON responses. Streaming is available for chat completions.
GET/v1/models
Enumerate hosted models with their public aliases.
POST/v1/chat/completions
Stream or fetch assistant replies using chat-style prompts.
POST/v1/completions
Generate classic text completions with temperature and stop controls.
POST/v1/embeddings
Produce vector embeddings optimised for semantic search and clustering.
POST/v1/responses
Unified endpoint that auto-falls back between chat and completion styles.
Security posture
Access is authenticated with scoped tokens and enforced quotas. Tokens can be disabled instantly, inactive users are blocked from invoking APIs, and every request is logged for audit trails.
- • CORS locked to https://ai.kentwynn.com to avoid rogue browser calls.
- • Quota enforcement guards both lifetime and daily token expenditure.
- • Admin dashboards require JWT auth and explicit account activation.
Sample JSON response
{
"choices": [
{
"message": {
"role": "assistant",
"content": "Kent Wynn AI keeps your LLM stack private and predictable."
},
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 18,
"completion_tokens": 24,
"total_tokens": 42
}
}Tooling & SDK support
Kent Wynn is compatible with the OpenAI ecosystem. Point existing SDKs at the Kent Wynn base URL or integrate with LangChain to orchestrate tools and agents.
LangChain
Use `langchain-openai` with a custom `base_url` to plug Kent Wynn models into existing chains, agents, and RAG pipelines.
View docs →OpenAI SDK drop-in
Point the official OpenAI Node/Python SDKs at our domain by setting `baseURL` and keep the rest of your integration unchanged.
View docs →curl-first
Prefer raw HTTP? Demo endpoints mirror OpenAI wire formats so you can test with curl or request libraries before coding.
View docs →LangChain ChatOpenAI example
Configure the LangChain OpenAI adapter with the Kent Wynn base URL to send prompts through the hosted stack.
from langchain_openai import ChatOpenAI
llm = ChatOpenAI(
base_url="https://api.kentwynn.com/v1",
api_key="YOUR_TOKEN",
model="kentwynn/reasoning",
extra_body={"tool_choice": "auto"},
)
response = llm.invoke(
"Give me two marketing angles for a private Kent Wynn deployment."
)
print(response.content)Ready to build on Kent Wynn AI?
Keep everything on your hardware while offering a polished developer experience. Set quotas, monitor usage, and deliver private LLM features in minutes.