GLM-5.3

DeepSeek-V4.1-Flash-1017 posts 77.0 on DeepSWE v1.1 against Opus 5's 74.0, with a 1M-token context and MIT-licensed weights. It still trails Opus 5 on Terminal-Bench 4.0 (42.3 vs 51.8) and ProgramBench (29.1 vs 37.0).

Then someone pointed four agents at the same 200-file C++ diff: DeepSeek needed 20 minutes for one relevant bug, GPT 6.1 Sol needed 4 and a half for two.

GLM-5.3 is presented as a frontier coding model with emergent cyber capabilities. HN readers liked the reported numbers but were cautious about turning release benchmarks into deployment decisions.

The sharper disagreement concerned access: if attackers can use capable models, defenders may need comparable tools without an approval process that only large labs can navigate.