# Cognition releases FrontierCode 2.0, a coding benchmark where top model scores 45%

> Original title: FrontierCode 2.0

- Source: Cognition Blog (Devin, Windsurf)
- Published: 2026-10-27T17:00:00.000Z
- Bad Influence: https://badinfluence.news/items/cz3amvu5stqxq1dijknp2x7np
- Original: https://cognition.com/blog/frontier-code-2.0

## Summary

Cognition introduces FrontierCode 2.0, a coding benchmark built with more than 20 open-source maintainers across 55% more repositories. The top-performing model, Claude Opus 5.5, scores 45.5%, down from 54.6% on FrontierCode 1.1. Thoroughness criteria, which test unstated correctness requirements, account for the largest share of blockers and have the highest failure rate at every reasoning effort level.

## Why it matters

The post explains which blocker categories drive scores and how reasoning effort changes results, giving a concrete view of where coding agents still fail.