Thinking Machines Lab · MoE agentic LLM

Inkling 2 Small

Inkling 2 Small is the first model of Thinking Machines Lab's second generation. It is a Mixture-of-Experts model with 594B total parameters and 17B active, accepts image inputs and supports variable effort. The lab reports a score of 31 on the Artificial Analysis Intelligence Index, well above its earlier models, with gains in agentic coding, vision, knowledge work and reasoning. It was trained with RL on hundreds of thousands of long-horizon coding tasks, using an agentic grader as well as unit-test rewards. On the PMPP-Hard GPU kernel benchmark it solved a mean of 54.25 of 69 tasks, and in a Tinker run it trained Qwen3.5-4B for time-series forecasting over 4.5 hours at a cost of $426, more than doubling the forecasting score. It is available on Tinker (Playground, fine-tuning and sampling) with closed weights for now; the lab intends to release the weights under Apache 2.0 the following week, after external safety testing and red-teaming.