# @davidad — 2026-02-26

♥824 ↻74 · https://x.com/davidad/status/2027075450481099153

Today you can download a 27B-parameter LLM that is generally smarter than o3, Sonnet 4, Grok 4, or DeepSeek’s 685B. On the one hand, this is terrifying. On the other hand, there’s never been a better time to do some artificial neuroscience and figure out how these things tick! https://t.co/7SWotUlwBV

tags: author:davidad, kind:tweet, model:claude-sonnet-4, model:grok-4, model:o3, on:claude-sonnet-4, year:2026
cited on: _dossiers/claude-sonnet-4.md, claude-sonnet-4