Lenovo's TianxiCode agent tops SWE-bench-Live Lite leaderboard at 71% resolution
Overview
Lenovo's TianxiCode coding-agent framework, paired with DeepSeek-v4.1-Flash, ranked first on the SWE-bench-Live Lite leaderboard with a 71% issue resolution rate, according to QbitAI (量子位).
The report says the result passed official Verified review. SWE-bench-Live is a software engineering benchmark built around real development environments.
The framework's approach, as described in the report, combines multi-hop retrieval, autonomous planning with multi-turn tool calling, and test-driven self-correction. The report states that Lenovo plans to apply it to its AI hardware products.
Written by AI from the articles below · updated Oct 9, 6:02 AM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
- QbitAILenovo's TianxiCode Agent Tops SWE-bench-Live Lite Leaderboard at 71%
AILenovo's TianxiCode, paired with DeepSeek-v4.1-Flash, ranked first on the SWE-bench-Live Lite leaderboard with a 71% issue resolution rate and passed official Verified review. The framework combines multi-hop retrieval, autonomous planning with multi-turn tool calling, and test-driven self-correction, and will be applied to Lenovo AI hardware products.
Heat trend
Not enough continuous observations to show a trend yet.