Gemini 4 Neon leads the Models Engineering Index at 69.5
Original title: Gemini 4 Neon is the new #1 on the Models Engineering Index
Anthropic's Fable 5.2 is second at 65.8 and Opus 5.5 third at 63.1 on the engineering-task index.
Mei, the Models Engineering Index, ranks language models on practical engineering tasks and also reports performance relative to token cost. Google's Gemini 4 Neon is listed first with a score of 69.5. The published table places Anthropic models in second and third, with OpenAI's 6.2 Astra fourth at 60.5.
Unconfirmed: Model names on the page are abbreviated, and the scoring method is not fully described.
- Google's 4 Hydrogen and Anthropic's Sonnet 5.5 both score 57.2.
- OpenAI's 6.2 Sol is seventh with a score of 56.9.
- Meta's Wonder Full ranks eighth at 49.3 on the index.
- The lowest listed score is OpenAI's 6 Luna at 34.8.
From the original article
1 min read at git-aarvin.github.io
Mei —or, the Models Engineering Index— calculates language models' practical performance on engineering tasks. 30405060704 Neon69. 5Fable 5.
By hn-aarvin · git-aarvin.github.io
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.