Top average (agent and edit) LiveSWEBench score by EOY2025?
2
1kṀ550
Dec 31
63.4 points
expected
79%
Above 50
69%
Above 60
57%
Above 70
45%
Above 80
27%
Above 90

LiveSWEBench (https://liveswebench.ai/) is a benchmark designed to evaluate the software engineering capabilities of AI agent applications.

This question ask about top average score in "Agentic Programming" AND "Target Editing" combined. Top score at 1 April 2025 is 47.83 (SWE-Agent with Claude Sonnet 3.7).

Will be judged according to the official leaderboard.

Get
Ṁ1,000
to start trading!
© Manifold Markets, Inc.Terms + Mana-only TermsPrivacyRules