Artificial Analysis launches FrontierCode 1.1 leaderboard measuring maintainer-ready coding changes

What happened

Artificial Analysis launched its FrontierCode 1.1 leaderboard, evaluating models on Cognition’s private coding benchmark using the 100-task Main set and its Stirrup open source reference harness. Grading combines unit tests with maintainer-oriented criteria; failing any blocking check yields zero. Claude Opus 5.5 (max with fallback) leads at 52%, with blockers passed in 58% of attempts; MiMo-V2.6-Pro leads open weights models at 45%, while GPT-6.1 Sol (max) scores 49% at $0.81 per task, matching Claude Fable 5.1 (max with fallback).

The story so far

Artificial Analysis launched a FrontierCode 1.1 leaderboard, with Claude Opus 5.5 (max with fallback) leading at 52% and MiMo-V2.6-Pro leading open-weights models at 45%. GPT-6.1 Sol (max) scored 49% at $0.81 per task, matching the score of Claude Fable 5.1 (max with fallback). The leaderboard uses the 100-task Main set from Cognition’s private coding benchmark and the Stirrup open-source reference harness. Artificial Analysis said the benchmark tests whether models can complete coding tasks in a way that maintainers of real-world repositories would accept for merging. Grading combines unit tests with maintainer-oriented criteria, and failing any blocking check gives an attempt a score of zero. Claude Opus 5.5 passed those blocking checks in 58% of attempts.

Keep reading

Trending now

1. 1Meta Releases Muse Spark Open With Mac and Cloud Support, Topping US Open-Weight Index
2. 2Google Launches Gemini 4 Argon Globally, Expands Paid Access as Benchmark Results Vary
3. 3OpenAI Updates GPT-6 Astra Pro in ChatGPT With Parallel Agents and Faster Responses
4. 4Meta’s New Business Terms Raise Dispute Over Website Crawling and AI Training Rights