Hacker News

Tencent Releases and Open-Sources Tencent Hy4 Preview

33 points by shenli3514 2 hours ago | 6 comments

Hy4 apparently has ludicrous traction on OpenRouter already (https://openrouter.ai/tencent/hy4-preview), with trillions of tokens generated in a couple days: more than GLM 5.3 in a week. That said, it's relatively cheap with a 5% cache cost when everyone is still doing 10%/20% cache costs, so Hy4 may be more compelling.

martinald 5 minutes ago

I wrote about this a couple of weeks ago. It's actually often the biggest cost and it tends to be hidden away on most platforms!

https://martinalderson.com/posts/watch-out-for-cache-read-co...

Btw I still haven't came across any decent model that is <$0.01/MTok cache costs apart from deepseek thru their official API (even with the price increases).

Seems like a bit of an opportunity for someone to take - drop cache read costs significantly.

Dinux 3 minutes ago

Which explains why almost none of my request go though

cyanydeez 40 minutes ago

i'd be curious if openrouter is just being gamed by these publishers by paying for the exposure.

wouldn't trust they dont do Capitalism like the rest of the AI field.

npn 19 minutes ago

[flagged]

vcryan 22 minutes ago

I used Hy3 quite a bit for the type of tasks it was suited for. Excited about this. My one concern over Hy3 was speed. In theory, it could be served much faster as a smaller model but it was relatively slow everywhere I could get it (including from Tencent directly) but also several other inference providers.

Topfi 15 minutes ago

In my evals, I saw an unprecedented jump between preview and final release on Hy3, from unusable to competitive. Did you see similar in preview vs release version?