Welcome Inkling 2 Mini, Cutting-edge Performance for your Local
Back to Articles a]:hidden"> Welcome Inkling 2 Mini, Cutting-edge Performance for your Local Setup Published November 20, 2026 Update on GitHub Upvote 15 +9…
Summary
Back to Articles a]:hidden"> Welcome Inkling 2 Mini, Cutting-edge Performance for your Local Setup Published November 20, 2026 Update on GitHub Upvote 15 +9 merve merve Follow Pedro Cuenca pcuenq Follow Aritra Roy Gosthipaty ariG23498 Follow Guarin Flück guarin Follow:last-child]:mb-0"> TLDR; Inkling 2 Mini is a new looped transformer model. It’s a multimodal Mixture-of-Experts with ~26B physical total and 3B active parameters per token. It uses a neat trick to increase capacity: some layers are visited in a loop that repeats three times, which makes it work like it had more parameters. It supports a 1M-token context size, accepts images and text inputs.
Positives
- It uses a neat trick to increase capacity: some layers are visited in a loop that repeats three times, which makes it work like it had more parameters.
Risks & concerns
- Back to Articles a]:hidden"> Welcome Inkling 2 Mini, Cutting-edge Performance for your Local Setup Published November 20, 2026 Update on GitHub Upvote 15 +9 merve merve Follow Pedro Cuenca pcuenq Follow Aritra Roy Gosthipaty ariG23498 Follow Guarin Flück guarin Follow:last-child]:mb-0"> TLDR; Inkling 2 Mini is a new looped transformer model.
- A DSpark drafter is also included for additional speed-up at a small memory cost.