Inkling 2 Small debuts at 31 on AA Intelligence Index, fine-tunes far faster than rivals
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 a full weight release planned.
A developer note highlights the key improvement: the original Inkling-Small was harder to post-train out of the box than Chinese models like Qwen, and Inkling 2 Small specifically addresses this — on a financial-document summarization task it reaches Inkling Small's final performance in just 75 training steps and finishes well ahead, beating GLM 5.3 Flash, DeepSeek V4.1 Flash and Qwen 3.5 397B on finetunability.
Related event: Thinking Machines Releases Inkling 2 Small, Open Weights Coming Next Week(42 posts)→
More from Models
- AI Cyber Skill Doubles Every 4-5 Months, Task Horizon Grows 52x in a Year — FinanceYF5 · 2026-11-06
- John Schulman Releases Inkling 2 Small With Detailed Training and Alignment Notes — ziqiao_ma · 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
- Qwen3.8-Flash-Next FP4 quantized model trends on Hugging Face — autotrust · 2026-11-06