OpenAI launches GPT-6.2 Astra and Sol
Astra's API price drops 20%
Models
Open weights · runs on a laptop
Written with AI assistance, reviewed by the ACN desk.
“Today we are releasing Inkling 2 Mini, a fast, powerful model that’s small enough to run on your laptop.”
Thinking Machines Lab has released Inkling 2 Mini, an open-weights model small enough to run on a laptop. The lab’s announcement says the model has a total memory footprint of about 20GB.
Inkling 2 Mini is a looped transformer, meaning some layers run repeatedly. Its two middle blocks of layers run three times for every token. The lab says it has 26.4 billion total parameters and 3.4 billion active. Thinking Machines says looping lets the model reach the right answer with fewer tokens, especially on agentic coding tasks.
The full weights are on Hugging Face, and the model is available on Tinker for inference and customization. The lab says one reason for the release is to let the community research looped transformers. It also says Inkling 2 Mini is not the strongest small model for multi-turn chat.
Artificial Analysis says the model scores 20 on its Intelligence Index. By its count, Inkling 2 Mini matches Qwen3.8 27B with about 5 times fewer active parameters. The firm says it uses about 38% more output tokens than Qwen3.8 27B for the same score. It is released under an Apache 2.0 license.
“Today we are releasing Inkling 2 Mini, a fast, powerful model that’s small enough to run on your laptop.”
“Inkling 2 Mini matches Qwen3.8 27B (xhigh) and Nanbeige4.3-33B on the Intelligence Index”
Sponsor
Spot an error? Email corrections@aicodingnews.si. Fixes are logged on the corrections page.
Sponsor
Sponsor
Astra's API price drops 20%
For long-running agent work · in GitHub Copilot the same day
MAI Code 2 and Code 2 Flash · in GitHub Copilot
In ChatGPT, Codex, the API and GitHub Copilot
Sponsor
Sponsor