← November 20, 2026 briefing · added in the 16:00 KST update

Gemini 4 Neon follow-up: tops engineering benchmark 'Models Engineering Index' at 68.0, ahead of Fable 5.2 (65.3) and GPT-6.2 Astra (61.5) (per community posts)

Announced November 20, 2026

Verified (fact)

Google's Gemini 4 Neon ranked first with 68.0 points on the Models Engineering Index (MEI), a public leaderboard comparing how well language models handle real engineering tasks. Anthropic's Claude Fable 5.2 placed second (65.3) and OpenAI's GPT-6.2 Astra third (61.5). Gemini 4 Carbon (61.4) and Claude Opus 5.5 (61.1) followed. The leaderboard lists 20 models in total, including models from Alibaba, Moonshot, Meta, and DeepSeek. According to a GeekNews summary, the evaluation tasks and scoring method cannot be confirmed from the published text alone. The leaderboard is hosted on a personal GitHub Pages site and received 182 points and 105 comments on Hacker News.

Why it matters (interpretation)

The result shows Gemini 4 Neon beating Fable 5.2 and GPT-6.2 Astra on an external engineering evaluation one day after its release. However, since it is run by an individual and the methodology is not public, it should be treated only as one reference point.

Confidence low · official source pending

Sources

Story thread

  1. 2026-11-19 Google DeepMind unveils Gemini 4 Neon, generally available in GitHub Copilot on launch day
  2. 2026-11-19 Gemini 4 Neon follow-up: analysis projects Agent Duel Elo of 1,066, near Opus 5.5's 1,078; ~2.8x Carbon's speed on OpenRouter, pricing more confusing (community)
  3. 2026-11-20 Gemini 4 Neon follow-up: tops engineering benchmark 'Models Engineering Index' at 68.0, ahead of Fable 5.2 (65.3) and GPT-6.2 Astra (61.5) (per community posts)

More about Google