Tiny open-source decision model beats Opus 5 and GPT-5.6

Name: FLock THIS/THAT Model 1.2: tiny open decision model beats Opus 5 + GPT-5.6
Uploaded: 2026-10-13T00:00:00+05:30
Description: FLock.io THIS/THAT Model 1.2 (built on Qwen3.5-2B, 1.88B params, MIT) scores 0.878 accuracy on 1,710 public questions in 31.4 ms with zero output tokens — ahead of Claude Opus 5 (0.834) and GPT-5.6 (0.816). Runs on laptops and phones.

Why it matters

FLock.io THIS/THAT Model 1.2 (built on Qwen3.5-2B, 1.88B params, MIT) scores 0.878 accuracy on 1,710 public questions in 31.4 ms with zero output tokens — ahead of Claude Opus 5 (0.834) and GPT-5.6 (0.816). Runs on laptops and phones.

Read the source: FLock (opens in a new tab)

Watch on YouTube Watch on Instagram

All hits from 13 Oct 2026 More Open Models Get the daily hits by email