GLM-5.1: Towards Long-Horizon Tasks • Hacker News Board

Re: GLM-5.1: Towards Long-Horizon Tasks

#161

Post by kcb » Wed, Apr 08, 2026, 2:18 AM UTC

deminature wrote: post #131

kcb wrote: post #119

Earlier quoted context omitted.

What benefit is there to dropping $50k on GPUs to run this personally besides being a cool enthusiast project?

Intel has just released a high VRAM card which allows you to have 128GB of VRAM for $4k. The prices are dropping rapidly. The local models aren't adapted to work on this setup yet, so performance is disappointing. But highly capable local models are becoming increasingly realistic. https://www.youtube.com/watch?v=RcIWhm16ouQ

That's 4 32GB GPUs with 600GB/s bw each. This model is not running on that scale GPUs. I think something like 96GB RTX PRO 6000 Blackwells would be the minimum to run a model of this size with performance in the range of subscription models.

Re: GLM-5.1: Towards Long-Horizon Tasks

#162

Post by foopod » Wed, Apr 08, 2026, 2:28 AM UTC

johnfn wrote: post #48

GLM-5.0 is the real deal as far as open source models go. In our internal benchmarks it consistently outperforms other open source models, and was on par with things like GPT-5.2. Note that we don't use it for coding - we use it for more fuzzy tasks.

It really bothers me that people refer to open weight models as being open source. They fundamentally aren't and are more akin to freeware than anything else.

Re: GLM-5.1: Towards Long-Horizon Tasks

#163

Post by stingraycharles » Wed, Apr 08, 2026, 2:30 AM UTC

simonw wrote: post #105

Not only did this one draw me an excellent pelican... it also animated it! https://simonwillison.net/2026/Apr/7/glm-51/

Surely at this point it’s part of the training set and the benchmark has lost its value?

Re: GLM-5.1: Towards Long-Horizon Tasks

#164

Post by sunir » Wed, Apr 08, 2026, 2:42 AM UTC

zozbot234 wrote: post #152

jimmaswell wrote: post #120

Earlier quoted context omitted.

No killer product? Coding assistants and LLM's in general are the single most awe-inspiring achievement of humanity in my lifetime, technological or otherwise. They've already massively improved my and others' lives and they're only going to get better. If pre and post industrial revolution used to be the major binary delineation of our history, I'm fairly confident it will soon be seen as pre and post AI instead.

Coding assistants are currently quite hard to run locally with anything like SOTA abilities. Support in the most popular local inference frameworks is still extremely half-baked (e.g. no seamless offload for larger-than-RAM models; no support for tensor-parallel inference across multiple GPUs, or multiple interconnected machines) and until that improves reliably it's hard to propose spending money on uber-expensive h…

Computers get better and cheaper. That’s not a forever problem.

Re: GLM-5.1: Towards Long-Horizon Tasks

#165

Post by sunir » Wed, Apr 08, 2026, 2:43 AM UTC

eldenring wrote: post #123

dvt wrote: post #118

Every single day, three things are becoming more and more clear: (1) OpenAI & Anthropic are absolutely cooked; it's obvious they have no moat (2) Local/private inference is the future of AI (3) There's *still* no killer product yet (so get to work!)

I don't see how its possible to think this. AI coding assistants are some of the most useful technologies ever created, and model quality is by far the most important thing, so I doesn't make sense why local inference would be the path forward unless something fundamentally changes about hardware.

The hardware will change. We know that.

Re: GLM-5.1: Towards Long-Horizon Tasks

#166

Post by marcus_holmes » Wed, Apr 08, 2026, 2:45 AM UTC

kcb wrote: post #158

marcus_holmes wrote: post #145

Earlier quoted context omitted.

Why would anyone need more than 640Kb of memory?

Exactly the point though. In the 640KB days there was no subscription to ever increasing compute resources as an alternative.

Well, there kinda was - most computing then was done on mainframes. Personal / Micro computers were seen as a hobby or toy that didn't need any "serious" amounts of memory. And then they ate the world and mainframes became sidelined into a specific niche only used by large institutions because legacy.

I can totally see the same happening here; on-device LLMs are a toy, and then they eat the world and everyone has their own personal LLM running on their own device and the cloud LLMs are a niche used by large institutions.

Re: GLM-5.1: Towards Long-Horizon Tasks

#167

Post by AlienRobot » Wed, Apr 08, 2026, 2:48 AM UTC

dvt wrote: post #118

Every single day, three things are becoming more and more clear: (1) OpenAI & Anthropic are absolutely cooked; it's obvious they have no moat (2) Local/private inference is the future of AI (3) There's *still* no killer product yet (so get to work!)

I was trying to use Claude.ai today to learn how to do hexagonal geometry.

Every time I asked a question it generated an interactive geometry graph on the fly in Javascript. Sometimes it spent minutes compiling and testing code on the server so it could make sure it was correct. I was really impressed.

Anyway I couldn't really learn anything since when the code didn't work I wasn't sure if I had ported it wrong or the AI did it wrong, so I ended up learning how to calculate SDF and pixel to hex grid from tutorials I found on google instead.

Re: GLM-5.1: Towards Long-Horizon Tasks

#168

Post by bottlepalm » Wed, Apr 08, 2026, 2:50 AM UTC

dvt wrote: post #118

Every single day, three things are becoming more and more clear: (1) OpenAI & Anthropic are absolutely cooked; it's obvious they have no moat (2) Local/private inference is the future of AI (3) There's *still* no killer product yet (so get to work!)

This has got to be bait..

1) OpenAI and Anthropic are killing it, and continue to do so, their coding tools are unmatched for professionals.

2) Local models don't hold a candle to SOTA models and there's nothing on the horizon that indicates that consumers will be able to run anything close to what you can get in a data center.

3) Coding is a killer product, OpenAI and Anthropic are raking in the cash. The top 3 apps are apps in the app store are AI. Everyone who knows anything is using AI, every day, across the economy.

Re: GLM-5.1: Towards Long-Horizon Tasks

#169

Post by stingraycharles » Wed, Apr 08, 2026, 2:52 AM UTC

pbhjpbhj wrote: post #109

QuantumNomad_ wrote: post #92

Earlier quoted context omitted.

Three hour coffee break while the LLM prepares scaffolding for the project.

Like computing used to be. When I first compiled a Linux kernel it ran overnight on a Pentium-S. I had little idea what I was doing, probably compiled all the modules by mistake.

I remember that time, where compiling Linux kernels was measured in hours. Then multi-core computing arrived, and after a few years it was down to 10 minutes.

With LLMs it feels more like the old punchcards, though.

Re: GLM-5.1: Towards Long-Horizon Tasks

#170

Post by kcb » Wed, Apr 08, 2026, 2:55 AM UTC

marcus_holmes wrote: post #166

kcb wrote: post #158

Earlier quoted context omitted.

Exactly the point though. In the 640KB days there was no subscription to ever increasing compute resources as an alternative.

Well, there kinda was - most computing then was done on mainframes. Personal / Micro computers were seen as a hobby or toy that didn't need any "serious" amounts of memory. And then they ate the world and mainframes became sidelined into a specific niche only used by large institutions because legacy. I can totally see the same happening here; on-device LLMs are a toy, and then they eat the world and everyone has the…

The difference is computers post text terminal are latency and throughput dependent to the user. LLMs are not particularly.