Thinky open-sources variable-loop model: more looping passes boost long-context but not knowledge benchmarks
simonguozirui · x · 2026-11-21
Thinky Machines released TML-d24-varloop-research, a variable-loop companion to Inkling 2 Mini (Loopling). Experiments show long-context benchmarks (GraphWalks BFS, MRCR v2) improve with more looping passes, while knowledge-intensive benchmarks (AA-Omniscience, SimpleQA Verified, SimpleVQA, WorldVQA) don't. Inkling 2 Mini outperforms July's Inkling 1.0 at just 2.7% of its size.
More from Models
- Gemini 4 Neon locked to Pro and Ultra paid subscribers, leaving Plus users out — giffmana · 2026-11-21
- xAI staff responds to Grok bot computer failures, escalates support ticket — mattyp · 2026-11-21
- Looped transformers: more looping boosts reasoning but with diminishing returns past 3 passes and fewer CoT tokens — nrehiew_ · 2026-11-21
- Team releases TML-d24-varloop-research on Hugging Face to study looping's effect on CoT monitorability — nrehiew_ · 2026-11-21
- Full sub-benchmark breakdown for Inkling 2 Mini published by Artificial Analysis — ArtificialAnlys · 2026-11-21
- Inkling 2 Mini costs ~6x less and runs ~4x faster per task than Qwen3.8 27B, AA finds — ArtificialAnlys · 2026-11-21