Thinking Machines ships Inkling 2 Mini: 26B looped transformer that runs on a laptop
_onionesque · x · 2026-11-22
Thinking Machines released Inkling 2 Mini, a looped transformer with just 26B parameters — a response to feedback that Inkling 2 Small at 594B was too big. The team says it's small enough to run on a laptop yet capable of agentic work, scoring 20.1 on AA Index. Benchmark charts show it on the Pareto frontier of both DeepSWE v1.1 and SWE-bench Pro v2 Hard (51 tasks), matching or beating Ripple-65B (Preview) and Qwen3.8-27B on task resolution at roughly an order of magnitude lower theoretical tokens per second.
More from coding & agent
- Agentic Benchmark: Gemini 4 Hydrogen Costs $0.37 Per Task vs Neon's $0.84 — Money-String4165 · 2026-11-22
- Uncensored Inkling-2-Mini (26B MoE) runs locally on 16GB, scores 77.9% on SWE-bench Pro Hard — udmrzn · 2026-11-22
- Dev sends 7 PRs to open-source coding agent Shelley after switching to Gemini — swaroopch · 2026-11-22
- Databricks previews agent platform that lets you git-branch your entire agent — matei_zaharia · 2026-11-22
- Taste MCP reins in AI's tendency to bloat UI with unneeded text and elements — annetgriffin · 2026-11-22
- Daniel Lemire: why you should write your own pull-request messages — lemire · 2026-11-22