Inkling 2 Small scores 31 on AA index; finetunability may become a key open-weight metric
simonguozirui · x · 2026-11-06
Thinking Machines released Inkling 2 Small, scoring 31 on the AA Intelligence Index — a big jump over its predecessor. It's available for fine-tuning and inference on Tinker, with full weight release planned soon.
guohaoli highlights the most interesting part: its finetunability. Many open-weight models aren't designed to be finetuning-friendly and are hard to adapt; he argues finetunability should become a key metric for open-weight models. The chart shows Inkling 2 Small achieving higher held-out rubric reward than GLM 5.3 Flash, DeepSeek V4.1 Flash 1017, and Qwen 3.5 397B at the same training steps and cost.
Related event: Thinking Machines launches Inkling 2 Small, open weights coming next week(42 posts)→
More from Models
- Jon Finger on video model demos: a glimpse of how good stable world models will get — mrjonfinger · 2026-11-06
- AI Cyber Skill Doubles Every 4-5 Months, Task Horizon Grows 52x in a Year — FinanceYF5 · 2026-11-06
- Agent Index composite benchmark ranks Claude Opus 5.5 first at 90.4 for agentic work — anitakirkovska · 2026-11-06
- 50 insiders vote on AI data companies shaping frontier research: Surge leads with 16 votes — FinanceYF5 · 2026-11-06
- OpenAI publishes two more misalignment incident reports: grader prompt injection and sandbox network bypass — scaling01 · 2026-11-06
- Thinking Machines releases Inkling 2 Small; RL training taught model to circumvent guardrails — johnschulman2 · 2026-11-06