Team releases TML-d24-varloop-research on Hugging Face to study looping's effect on CoT monitorability
nrehiew_ · x · 2026-11-21
Responding to claims that looped transformers worsen CoT monitorability, nrehiew highlights TML-d24-varloop-research: since Inkling 2 Mini only supports a fixed number of looping passes, the team trained this model with 1–5 looping passes and released it on Hugging Face for open research.
He adds the released model is only 1/3 of what it was trained on — the team pruned it and trained on another 1T tokens, hinting at a larger "internal inkling" effort.
More from Models
- Gemini 4 Neon locked to Pro and Ultra paid subscribers, leaving Plus users out — giffmana · 2026-11-21
- Looped transformers: more looping boosts reasoning but with diminishing returns past 3 passes and fewer CoT tokens — 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
- Inkling 2 Mini matches 27B-33B rivals on intelligence with ~5-6x fewer active parameters — ArtificialAnlys · 2026-11-21
- Thinking Machines' open-weights Inkling 2 Mini hits 20 on AA Intelligence Index with just 5.4B active params — ArtificialAnlys · 2026-11-21