News for hackers. Stuff that matters.

A comment on Addendum to GPT-5 system card: GPT-5-Codex. Parent comment

Re:Addendum to GPT-5 system card: GPT-5-Codex by vitorgrs on Monday September 15, 2025 @11:33PM (#45257760)
Gemini seems to be pretty awful as agentic coding. It always finish the task, and when I see the result, it just breaks my code.

Not sure the fault it's "doing bad code", I guess it's just not being good at being agentic. Saw this on Gemini CLI and other tools.

GLM, Kimi, Qwen-Code all behaves better for me.

Probably Gemini 3 will fix this, as Gemini 2.5 Pro it's "old" by now.

[ Parent | on HN ]
Re:Addendum to GPT-5 system card: GPT-5-Codex by faangguyindia on Tuesday September 16, 2025 @12:19AM (#45257979)
Gemini CLI is bad, model itself is really good.
[ Parent | on HN ]