Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Hy3

hy.tencent.com

Post reply on HN
Print view View original

31–40 of 125 posts

Re: Hy3

#31

Post by Miner49er »

minimaxir wrote: post #24

A month ago I wrote a blog post about how Hy3 was topping the OpenRouter rankings despite no one talking about it: https://news.ycombinator.com/item?id=48317294 As of today, it has fallen to 8/9th on the rankings. I don't see a reason where you would use this model over competitors. However, price economics are bit confusing, as currently the effective input price of Hy3 via OpenRouter is now the same as DeepSeek-hos…

I had to stop using it because I was getting rate limited like crazy. Probably why it has dropped. Seemed like they couldn't keep up with demand.

Re: Hy3

#32

Post by Catloafdev »

UncleOxidant wrote: post #16
Catloafdev wrote: post #9

Curious how people feel about this compared to DS4 Flash, given they are pretty close in size. Also curious how well it holds up to heavy quantization. DS4 Flash can currently run reasonably well on systems with ~96gb+ RAM, I wonder if Hy3 can compete there.

That's a 2-bit quant of DS4 flash. You're probably better off running Qwen3.6-27B at Q8.

For most coding or agentic tasks, Qwen 3.6 27B likely outperforms, yes.

For 'general intelligence', DS4 Flash seems to be a noticeable step up still.

Re: Hy3

#34

Post by wgd »

andai wrote: post #10
nshotton wrote: post #4

This model is shockingly small for how capable it is. its a little bit bigger than deepseekV4 flash but around as capable if not more on some benchmarks than V4 pro, i wouldnt be surprised if this becomes a popular local model.

I've been wondering about that. GLM-5.2 is also half the size of DeepSeek V4 Pro. (But costs roughly twice as much.) I looked into DeepSeek's architecture a little bit and the main focus was how can we save as much money as possible. They did a lot of cost cutting with the attention mechanisms. This allowed them to offer an insanely cheap price even on massive contexts, but seems to have come at the cost of performan…

The current Deepseek V4 Pro is still just their initial preview AFAIK, with the "real" model release rumored to come later this month. GLM-5.2 might be outperforming simply because it's had more post-training on top of the GLM-5 base.

Re: Hy3

#35

Post by thot_experiment »

I feel like I'm taking crazy pills with hy3, it's either benchmaxxed to hell and back or skill issue on my part but I'd rather use dense gemma. I don't think there's a single model that's wasted more of my time in recent memory.

Re: Hy3

#36

Post by ckocagil »

UncleOxidant wrote: post #16
Catloafdev wrote: post #9

Curious how people feel about this compared to DS4 Flash, given they are pretty close in size. Also curious how well it holds up to heavy quantization. DS4 Flash can currently run reasonably well on systems with ~96gb+ RAM, I wonder if Hy3 can compete there.

That's a 2-bit quant of DS4 flash. You're probably better off running Qwen3.6-27B at Q8.

Isn't Q8 way overkill these days? I see many graphs showing Q4 or Q5 having less than %1 deviation. Nvidia's NVFP4 Qwen quantization should be even better due to its better training methods.

Re: Hy3

#37

Post by mdgld »

ckocagil wrote: post #36
UncleOxidant wrote: post #16

Earlier quoted context omitted.

That's a 2-bit quant of DS4 flash. You're probably better off running Qwen3.6-27B at Q8.

Isn't Q8 way overkill these days? I see many graphs showing Q4 or Q5 having less than %1 deviation. Nvidia's NVFP4 Qwen quantization should be even better due to its better training methods.

It depends on model size I think, but yeah, from my understanding at ~30B and below Q6 or even Q4 will get you 95%+ of the way there

Re: Hy3

#38

Post by tarruda »

Catloafdev wrote: post #9

Curious how people feel about this compared to DS4 Flash, given they are pretty close in size. Also curious how well it holds up to heavy quantization. DS4 Flash can currently run reasonably well on systems with ~96gb+ RAM, I wonder if Hy3 can compete there.

> given they are pretty close in size

One thing that might not be obvious about about DSV4 is how much innovation the Deepseek team implemented in its architecture. When llama.cpp fully supports its lightning indexer, the full 1M context will only require about 6G of RAM. So even though they are similar in size, I believe Deepseek will be much more efficient in that regard.

> I wonder if Hy3 can compete there

Highly depends on how well Hy3 is resilient to quantization. DSV4 is useful even at 2-bit quants.

Re: Hy3

#39

Post by docheinestages »

What we really need is a breakthrough in inference or LLM architecture to allow running GLM-5.2-level models at the size of Qwen 3.6 27b or smaller on consumer devices like a 48GB Macbook Pro, and at least at 100 tokens/second. My hypothesis is that a smaller, less capable but faster model paired with a good harness can run for longer and brute force its way out to solve problems that the bigger models can one-shot.

Re: Hy3

#40

Post by cyanydeez »

docheinestages wrote: post #39

What we really need is a breakthrough in inference or LLM architecture to allow running GLM-5.2-level models at the size of Qwen 3.6 27b or smaller on consumer devices like a 48GB Macbook Pro, and at least at 100 tokens/second. My hypothesis is that a smaller, less capable but faster model paired with a good harness can run for longer and brute force its way out to solve problems that the bigger models can one-shot.

im more expecting the harness to be a literal LLM, Like how you put vibration dampeners on all kinds of mechanical structures
Post reply on HN