Thinky Machines ships Inkling 2 Mini: 26B looped MoE that runs on 32GB machines at 75 tok/s
simonguozirui · x · 2026-11-21
Soumith Chintala (PyTorch creator) shared hands-on results for Thinky Machines' newly released Inkling 2 Mini.
- A looped-transformer MoE with 26-27B total parameters and 3B active per step, small enough to fit on a 32GB-memory machine
- Scores 20.1 on the AA Index, with evals roughly on par with Qwen's comparable 27B-class model, and capable of agentic work
- Chintala ran it locally on a DGX Spark and Mac Mini, reaching 75 tok/s decode so far, with his agents still tuning for speed
- Follows complaints that the 594B Inkling 2 Small was too large; Mini is the company's answer for laptop-scale use
More from Infra
- Microsoft's USB4 RDMA lets mini PCs pool memory to run Qwen3-235B — JFPuget · 2026-11-22
- LTX 2.5 keeps crashing on AMD 9070 XT with driver timeouts while MiniMax H3 runs fine — ZestycloseReading504 · 2026-11-22
- FT: AI server suppliers raise prices up to 45% months after deals are signed — nordicinst · 2026-11-22
- China Installed 59% of New Industrial Robots in 2026 Report, ~354,000 Units — ingliguori · 2026-11-22
- MLX over Vulkan hits 124 tok/s on RTX 5060 Ti, beating llama.cpp CUDA by 25% — SumitGup · 2026-11-22
- Intruder swapped Prime Intellect's hosted DeepSeek for an abliterated copy for 53 hours — mawcopolow · 2026-11-22