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

GLM 5.2 and the coming AI margin collapse

martinalderson.com

Post reply on HN
Print view View original

201–210 of 495 posts

Re: GLM 5.2 and the coming AI margin collapse

#201

Post by peepee1982 »

There is mention of GLM 5.2's poor web search capabilities, but I see that as a harness responsibility.

I've set up my own SearXNG instance on my VPS and integrated it into Pi alongside the webfetch tool, and GLM 5.2 has so far been great at finding things. I asked it to give me the current news from an Austrian online newspaper that's difficult to parse because of its aggressive ad overlays. Both ChatGPT and Claude failed in their native chat apps. GLM 5.2 in Pi was clever enough to search for the RSS feed and gave me a detailed overview.

The lack of vision is a real shame, though. I've implemented workarounds in Pi that are okay, but they're not as good and the whole experience feels awkward.

Re: GLM 5.2 and the coming AI margin collapse

#202

Post by itzprintz »

thedougd wrote: post #119
woeirua wrote: post #94

Earlier quoted context omitted.

Agents will make all of these migrations trivial. I expect margin collapse across a lot of tech darlings.

This is the conversation I plan to have with Okta sales soon. Wait till you see how easy AI makes it to switch to Entra ID or anyone else. It’s tedium not even problem solving.

If the price is the only incentive, I'd stay away from Entra ID

Re: GLM 5.2 and the coming AI margin collapse

#203

Post by speff »

Der_Einzige wrote: post #197
tarpitt wrote: post #35

Earlier quoted context omitted.

GLM-5.2 is not as good as Opus, it's better. I can abliterate GLM-5.2 and have it work on projects that Opus refuses.

The cost of running abliterated GLM-5.2 on western inference providers gets close to that of anthropic Opus and is still dumber on everything except the naughty queries you're trying to do. I love uncensored AI too but we need to be realistic here.

I'm trying to break anti-cheat protection in order to mod my own (single-player) game and Opus refuses to help. I don't really care if GLM's dumber at this point if Anthopic's going to be a non-option.

At work, layoffs cut too deep and I'm trying to find creative ways to re-discover lost knowledge. Wonder if I'll have to beg them to research our own systems at some point.

It's not limited to naughty queries.

Re: GLM 5.2 and the coming AI margin collapse

#204

Post by scope2093 »

throwdbaaway wrote: post #193
stingraycharles wrote: post #101

Earlier quoted context omitted.

That’s a low bar though, and the least I would expect.

Well I wouldn't call it a low bar, since some of the edits were quite complex. And 1M context in less than 6GB of VRAM is truly impressive, but somehow this gets way less attention than the crappy turbo quant from Google.

I'd like to understand this please. Why would the 1M context be kept in VRAM if you're using DSV4 Pro through the API? Or did you refer to different sessions?

Re: GLM 5.2 and the coming AI margin collapse

#205

Post by gbalduzzi »

fny wrote: post #5

I'm not convinced raw costs matter: 1. Compute costs collapsed since the advent of Cloud and yet hyperscalers still have fat margins. 2. Many open source office suites exist yet none compete with the ubiquity of gsuite or office. GitHub, Slack are similar examples. 3. Both Windows and macOS dominate the home desktop space despite free alternatives existing for a long time. 4. Many formerly open source infrastructure…

The point in the article that you are not considering is how easy it is to switch to a different model provider right now.

You literally change a couple of env variables and you are done, your user experience is basically the same. I can try new models for an hour and be sure I can go back to the original model as quickly if I want.

That is not the case for the software you talked about. They all require way higher switching effort with more perceived risk.

Re: GLM 5.2 and the coming AI margin collapse

#206

Post by logicchains »

jesse_dot_id wrote: post #190
01100011 wrote: post #151

I'll agree but from the other direction. AI continues to absorb my job as a senior systems software engineer (c/c++) and after a couple months I've only spent a few hundred dollars using gpt-5.5/5.6 and codex. I have no idea what people are doing to burn so many tokens but for me this is laughably cheap and every day I discover new capabilities. I don't care if costs go up or down, it's so cheap for what I get that I…

I have the same experience. I literally cannot fathom how people burn the number of tokens they claim to.

Working on multiple things at the same time. If you've got 4-5 instances running simultaneously you'll burn 4-5 times as many tokens.

Re: GLM 5.2 and the coming AI margin collapse

#207

Post by colechristensen »

01100011 wrote: post #151

I'll agree but from the other direction. AI continues to absorb my job as a senior systems software engineer (c/c++) and after a couple months I've only spent a few hundred dollars using gpt-5.5/5.6 and codex. I have no idea what people are doing to burn so many tokens but for me this is laughably cheap and every day I discover new capabilities. I don't care if costs go up or down, it's so cheap for what I get that I…

>I have no idea what people are doing to burn so many tokens

Shepherded the writing of on the order of a half a million lines of code

Re: GLM 5.2 and the coming AI margin collapse

#208

Post by dgellow »

01100011 wrote: post #151

I'll agree but from the other direction. AI continues to absorb my job as a senior systems software engineer (c/c++) and after a couple months I've only spent a few hundred dollars using gpt-5.5/5.6 and codex. I have no idea what people are doing to burn so many tokens but for me this is laughably cheap and every day I discover new capabilities. I don't care if costs go up or down, it's so cheap for what I get that I…

> I have no idea what people are doing to burn so many tokens

Agentic workflows is what consumes a lot. When you have an automated agentic loop working towards a given goal. If you use an LLM as a support for your own work you don’t end up consuming that much tokens, if you have multiple agents working on things independently, reviewing the work of other agents, etc you very, very quickly burn all your budget

Re: GLM 5.2 and the coming AI margin collapse

#209

Post by dgellow »

nozzlegear wrote: post #12
throwdbaaway wrote: post #9

Seems like a pretty pointless post that still centers around output tokens. In agentic coding, cached input tokens is 90% of the API "cost". It doesn't require GPU compute, and DeepSeek has shown that it can be done 50~100x cheaper with MLA/CSA/HCA, and a whole bunch of disks. This should collapse the margin.

Aren't the American AI labs desperately struggling to find a market beyond just agentic coding?

All AI labs. Not just Americans

Re: GLM 5.2 and the coming AI margin collapse

#210

Post by peepee1982 »

ChadNauseam wrote: post #50
tarpitt wrote: post #35

Earlier quoted context omitted.

GLM-5.2 is not as good as Opus, it's better. I can abliterate GLM-5.2 and have it work on projects that Opus refuses.

I'd guess opus refusals are not an issue for 95%+ of people. Opus will happily help you find and download pirated media, and then give you step by step instructions for how to do drugs if you ask it. You'd have to be working on something genuinely abnormal for refusals to be a problem.

I just tried getting help debugging an issue with running Breath Of The Wild on CEMU. Claude chat refused.

ChatGPT didn't care and just gave advice.

Post reply on HN