Agent skills for TokenLab API integration, model picking, native endpoints, cost routing, and OpenAI-compatible migration
-
Updated
Oct 12, 2026 - Python
Agent skills for TokenLab API integration, model picking, native endpoints, cost routing, and OpenAI-compatible migration
MCP client configuration example for TokenLab
Vercel AI SDK 7 example for TokenLab
TokenLab provider for Vercel AI SDK using TokenLab's OpenAI-compatible endpoint
TokenLab model provider plugin for Hermes Agent
TokenLab examples for Chat Completions, Responses, Anthropic Messages, and Gemini
LlamaIndex Python example for TokenLab
Next.js AI chat starter for TokenLab with AI SDK 7
Lightweight Python SDK for TokenLab discovery, OpenAI-compatible APIs, and native endpoints
Lightweight JavaScript SDK for TokenLab discovery, OpenAI-compatible APIs, and native endpoints
TokenLab native-protocol model provider, multimodal tools, and async tasks for DeepSeek Harness
Importable n8n workflows for TokenLab model routing, async media polling, and OpenAI-compatible chat
Hosted TokenLab Model Explorer for ChatGPT Apps and Streamable HTTP MCP
TokenLab model provider plugin for OpenClaw with live catalog discovery
OpenAPI-generated MCP server for TokenLab model discovery, pricing, chat, native System One decisions, media, files, embeddings, webhooks and async tasks.
To associate your repository with the tokenlab topic, visit your repo's landing page and select "manage topics."