Inkling 2 Mini: Thinking Machines' 26B open model you can run on a laptop
Thinking Machines released Inkling 2 Mini, an Apache 2.0 looped MoE model of about 20GB that targets 32GB Macs and DGX Spark. Here is which quant fits your memory and what still breaks on launch day.
Thinking Machines released Inkling 2 Mini, an Apache 2.0 looped MoE model of about 20GB that targets 32GB Macs and DGX Spark. Here is which quant fits your memory and what still breaks on launch day.
Inkling 2 Mini: how to run Thinking Machines' 26B open model locally, and which quant fits your RAM
Inkling 2 Mini is Thinking Machines' Apache 2.0 looped MoE model: 26.4B parameters, 3.4B active, about 20GB. Which GGUF quant fits your memory, the commands for llama.cpp, MLX and vLLM, launch-day gaps, and how it compares with Qwen3.8-27B.