Inkling
Inkling is Thinking Machines’ multimodal reasoning model, built for deep analysis, coding, and tool-driven work across text, images, and documents. It uses a 975B-parameter mixture-of-experts architecture with 41B active parameters and a 256K-token context window.
Powered by a server managed by @thinkymachines. Learn more
Build with Inkling using the Poe API
Start by creating an API key, for use with any bot on Poe:
See the full documentation for comprehensive guidance on getting started.
More from Thinking Machines
Inkling