Cursor vs Continue vs Open WebUI vs LiteLLM
Choose where TKEN should sit in a developer workflow before routing IDE, chat UI, proxy or team-key traffic.
Developer resource hub
Build on the TKEN OpenAI-compatible API gateway
Start with one base URL, one API key, and OpenAI-route target pricing at one tenth of official OpenAI API list price where available. Use these guides for SDKs, agents, dashboards, coding tools, reliability patterns, and cost controls.
TKEN base URL
OpenAI-compatible
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.TKEN_API_KEY,
baseURL: "https://www.tken.shop/v1"
});
const result = await client.chat.completions.create({
model: "MiniMax-M2.7",
messages: [{ role: "user", content: "Run one smoke test." }]
});Base URL: https://www.tken.shop/v1
OpenAI routes target 1/10 official API list price
GPT-5.4 Mini from $0.075 / $0.45 per 1M tokens
OpenAI SDK-compatible request shape
Guides for tools, retries and cost guardrails
Start here
These pages help a new developer understand the gateway, create a key, run one request, and compare costs before moving into a full app or tool integration.
Overview
First request
SDK setup
OpenAI route target
1/10 official OpenAI API list price
Low-cost anchor
GPT-5.4 Mini: $0.075 / $0.45 per 1M
Live prices
SDK and tool setup
Endpoint preflight
Verify /v1, /models, selected model IDs, one chat request and safe evidence before SDK, proxy, agent or UI rollout.
MCP host gateway config review
Separate MCP server permissions from the OpenAI-compatible model route before connecting hosts, IDEs or agent runners to TKEN.
Open WebUI
Configure Open WebUI with the TKEN base URL and a gateway API key.
LiteLLM
Route LiteLLM requests through an OpenAI-compatible gateway while keeping a standard provider shape.
Cursor
Set an OpenAI-compatible base URL for coding workflows that support custom endpoints.
Continue
Use TKEN as an OpenAI-compatible provider in Continue for editor assistance.
MCP tool calling
Plan MCP and tool-calling flows with conservative compatibility checks and a clear smoke-test path.
API patterns
Use these pages when you are moving from a successful smoke test to a production workflow with streaming, embeddings, structured outputs and predictable failure handling.
Embeddings
Streaming
Retries
Reliability, security and cost control
API key rotation
Rotate keys and store secrets without exposing them in browser code, public repos, logs or support conversations.
LLM cost guardrails
Put model choice, max tokens, retry limits and recharge checks in place before production traffic grows.
OpenAI-route pricing guide
Use the 1/10 target price anchor for planning, then confirm the current model price on live /pricing before routing production traffic.
TKEN is an independent third-party API gateway. It is not an official
endpoint for OpenAI, Anthropic, Google, DeepSeek, Qwen, Kimi, MiniMax,
GLM or other model providers. OpenAI-model routes target one tenth of
official OpenAI API list price where available, but model availability,
pricing and compatibility can change. Live /pricing remains the source
of truth; run a smoke test before production use.
Use one request to verify your setup
Create an API key, set https://www.tken.shop/v1 as the base URL, test one chat completion, then confirm live pricing before moving a tool or app to TKEN.