GLM-5.2: Frontier Intelligence, Open Weights

Given that DeepSwe is one of the very few coding benchmarks worth taking a look at, this achieves rather excellent result at it (not far from opus 4.8).

From looking at the results and my own impression of 5.1 and other models, I think this is the best Chinese coding model by some non-insignificant margin.

▲

LaurensBER 9 hours ago | parent [-]

I've been very pleased with it's performance over the last few days.

It's definitely not near Opus 4.8 level but it's very impressive nonetheless and it does do design extremely well.

▲

ebbi 8 hours ago | parent [-]

> it does do design extremely well

Better than Opus?

▲

osti 4 hours ago | parent [-]

I don't know what people mean when they say design lol, is it for frontends?

	▲	ebbi 32 minutes ago \| parent [-]
		Yeah, that's what I mean anyway. Each model has certain design tropes it repeats everywhere, and some of them are very old-school or not really UI best practice. And then the more ambitious cases where you ask for a feature without being prescriptive with UI needs, the end result is sometimes atrocious with weird font use, colours, etc.

▲

fcanesin 6 hours ago | parent | prev | next [-]

Zhipu AI is founded by a superstar Tsinghua professor, did an IPO in January (Hong Kong stock exchange) hired half it's past research lab and it's stock is >10x since. This is not a "just distill Claude" thing.

▲

hadik 9 hours ago | parent | prev | next [-]

Blog post with more information including benchmarks: https://z.ai/blog/glm-5.2

▲

coolspot 5 hours ago | parent | prev | next [-]

Looks like their code subscription is sold out. Clicking on the "Subscribe" button just does nothing. Pay-as-you-go API works well.

▲

gnabgib 6 hours ago | parent | prev | next [-]

Discussion (765 points, 3 days ago, 489 comments) https://news.ycombinator.com/item?id=48518684

▲

dbbk 9 hours ago | parent | prev | next [-]

If I have a fully maxed out MacBook Pro, would it make sense to just switch from Opus 4.8 to this? I've never tried running local models for coding...

▲

entrope 7 hours ago | parent [-]

HuggingFace says this model has 753B parameters, which will need a lot more RAM than a maxed-out MacBook Pro. With 40B active parameters, running from SSD would need patience.

	▲	_aavaa_ 4 hours ago \| parent \| next [-]
		For an fp4 quantization it should fit with room to spare for KVCache
	▲	api 7 hours ago \| parent \| prev [-]
		I’ve wondered for a while if anyone is working on very wide channel parallel (kind of like RAID 0) SSD for this purpose. Couple that with a tensor processor and that would be interesting.

▲

Bucko1 10 hours ago | parent | prev | next [-]

I have already tested this and it is impressive.

▲

ChrisArchitect 6 hours ago | parent | prev [-]

non-x blog post: https://z.ai/blog/glm-5.2

And previously: https://news.ycombinator.com/item?id=48518684