VQV Signal

OPEN SOURCE WATCH TECHNICAL

Thinking Machines releases Inkling 2 Mini, a 26B open-weights model it says runs on a laptop

According to a post on X by @thinkymachines, Thinking Machines released Inkling 2 Mini, a 26B-parameter looped mixture-of-experts transformer under the Apache 2.0 license. The company says it runs on a laptop and scores 20.1 on the Artificial Analysis Index. The snippet also cites Unsloth saying it runs in 16GB of RAM, and Vals AI saying it ties Claude Opus 5.5 on CyberBench at under a quarter of the cost.

Source: Scobleizer News Pulse (X) · x.com Found via Robert Scoble’s Scobleizer News Pulse · a post on X, not independently verified Published 2026-11-20T21:15:00+00:00 Detected 2026-11-21T05:00:38+00:00
View original source

According to a post on X by @thinkymachines, Thinking Machines released Inkling 2 Mini, a 26B-parameter looped mixture-of-experts transformer under the Apache 2.0 license. The company says it runs on a laptop and scores 20.1 on the Artificial Analysis Index. The snippet also cites Unsloth saying it runs in 16GB of RAM, and Vals AI saying it ties Claude Opus 5.5 on CyberBench at under a quarter of the cost.

AI-assisted summary based on the listed source.

Thinking Machines released Inkling 2 Mini, a 26B-parameter looped mixture-of-experts transformer under Apache 2.0 that it says runs on a laptop and scores 20.1 on the Artificial Analysis Index. Unsloth says it runs in 16GB of RAM, and Vals AI says it ties Claude Opus 5.5 on CyberBench at under a quarter of the...

The claims point to a capable open-weights model that could run on personal hardware instead of only in data centers. Developers considering local deployment may want to test the model on their own tasks and machines, since the performance figures come from the company and third parties. Independent benchmark results and early adoption will show whether these claims hold up.

Signal Strength 88% Technical label WATCH Public Interest 49 Category OPEN SOURCE Reader Depth TECHNICAL

Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.

Public Interest components
Recognizable Entity Score 51 Practical Impact Score 36 Novelty Interest Score 94 Consequence Score 18 Curiosity Score 16 Shareability Score 62

VQV surfaced this signal because it is recent, relevant to Consumer AI, connected to Scobleizer News Pulse (X).