News for hackers. Stuff that matters.

A comment on Gemini 4 Hydrogen. Parent comment

Re:Gemini 4 Hydrogen by seizethecheese on Thursday November 12, 2026 @02:45PM (#50473111)
It's definitely not the case that small models only do well in benchmarks. We're working on a battle system on top of OpenCode, so have data for real people doign real work.

Right now DeepSeek Flash 4.1 Flash has a 64% win rate against GPT 6.1 Sol. Small data still, but it has 33 wins vs. 18 losses, which is confidence interval of 52% - 78% win rate.

[ Parent | on HN ]
Re:Gemini 4 Hydrogen by redox99 on Thursday November 12, 2026 @03:41PM (#50473944)
That's coding though. I was talking about general usage.
[ Parent | on HN ]
Re:Gemini 4 Hydrogen by seizethecheese on Thursday November 12, 2026 @03:57PM (#50474169)
Okay, makes sense, though I don’t see a huge difference in quality between Luna and Astra for basic chat usage.
[ Parent | on HN ]