---
format: "aidr-story-markdown/v1"
id: "f208142017e0a401209f4bc8b8a789962a6291ad5a5f42e786c748b0a3ddbc12"
canonical_url: "https://aidr.today/f2081420?lang=en"
title: "Leading AI Coding Models Score Below 50% on Cognition's New FrontierCode Benchmark"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-27T19:09:22.000Z"
category: "Models"
topics: ["coding"]
source_urls: ["https://marketbrief.now/ai/leading-ai-coding-models-score-below-50percent-on-cognitions-new-frontie-4fc2ac6c","https://huggingnews.com/ai/leading-ai-coding-models-score-below-50percent-on-cognitions-new-frontie-4fc2ac6c"]
summary: "Cognition released the FrontierCode 2.0 benchmark, which evaluates AI coding models on design choices rather than functional correctness, with top models scoring below 50%."
---

# Leading AI Coding Models Score Below 50% on Cognition's New FrontierCode Benchmark

> [Open the canonical story](<https://aidr.today/f2081420?lang=en>)

**Published:** 2026-10-27T19:09:22.000Z
**Category:** Models
**Topics:** coding

## Summary

Cognition released the FrontierCode 2\.0 benchmark, which evaluates AI coding models on design choices rather than functional correctness, with top models scoring below 50%\.

## Sources

- [Story source](<https://marketbrief.now/ai/leading-ai-coding-models-score-below-50percent-on-cognitions-new-frontie-4fc2ac6c>)
- [Story source](<https://huggingnews.com/ai/leading-ai-coding-models-score-below-50percent-on-cognitions-new-frontie-4fc2ac6c>)