Thinking Machines' Inkling 2 Mini lands on Ollama, runs locally on a 32GB Mac
ollama · x · 2026-11-21
Thinking Machines' small open-weights model Inkling 2 Mini is now available on Ollama (v0.40.17 pre-release), runnable fully locally via ollama run inkling-2-mini.
- Specs: 26B total parameters with only 5.4B active per token; middle layers loop three times per token for deeper reasoning instead of generating more tokens; runs on any Mac with 32GB+ memory
- Focus: built for coding, agentic workflows like computer use, design tasks requiring strong visual reasoning, and finance; supports vision, tools, and thinking
- Ollama also demoed the model playing 2048 with real-time decisions
More from coding & agent
- FrontierRSI benchmark: best AI scores only ~34% at improving AI, says Abundant — Madisonkanna · 2026-11-21
- Changing one MCP tool description sentence lifted correct tool selection from 19/55 to 45/55 — sparkignitefire · 2026-11-21
- Founder's rebuttal to McKinsey: enterprises need a first brain, not an AI second brain — vasuman · 2026-11-21
- Devs: LLMs absorb layered complexity until context explodes; tools and MCP are too heavy — dscape · 2026-11-21
- Inkling 2 Mini on Ollama: MLX build, one-command hooks into Claude Code and OpenCode — ollama · 2026-11-21
- Formal Verification Emerges as the Weapon Against AI Slop, Powering a New Startup Wave — Latent Space · 2026-11-21