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

GLM 5.2 vs. Opus

techstackups.com

Post reply on HN
Print view View original

231–240 of 367 posts

Re: GLM 5.2 vs. Opus

#231

Post by Der_Einzige »

scwoodal wrote: post #104
cultofmetatron wrote: post #19

I seriously dont' know all this big hullabaloo about one shot prompting. by definition, a single prompt wont' constitute the complexity of a software project. ergo, what you'll get is a series of assumptions made by the model based on preexisting code in its training corpus. I'd rather see a coding agent that can follow steps in a plan file to a T while following guardrails and adhering to the proper coding conventio…

> I'd rather see a coding agent that can follow steps in a plan file to a T while following guardrails and adhering to the proper coding conventions in the human reviewed spec. Guardrails/conventions should be enforced in linters, formatters, static analysis tooling; not specs/prompts.

Wrong, custom "specs" i.e. schemas, are literally all we have for "real" guardrails with LLMs.

https://developers.openai.com/api/docs/guides/structured-out...

Nothing else operates on the logprobs level and literally bans continuations that fail your schema.

Re: GLM 5.2 vs. Opus

#232

Post by newaccountman2 »

blanched wrote: post #230
newaccountman2 wrote: post #191

Earlier quoted context omitted.

[flagged]

This kind of hamfisted snark tends to make people take the actual and justified criticism of police less seriously.

If people were willing to take it seriously in the first place, then they wouldn't view it as "hamfisted snark"

Re: GLM 5.2 vs. Opus

#233

Post by Der_Einzige »

cicko wrote: post #54
jack_pp wrote: post #33

Earlier quoted context omitted.

This framing local LLMs as free is stupid. Basically pay 100+ months worth of API costs up front isn't free in the slightest. And it will be slower than non-local, your hardware will be outdated in 12 months and probably won't be able to run SOTA at anywhere near non-local speed in max 20 months

True. But there are other meanings of "free". I.e. nobody can say "from now on you no longer have access to model X because you're an asshole"

The ecosystem for inference is centralized around a few core projects, i.e. vLLM, sglang, and llamacpp.

If they decided to collude, they could absolutely say "from now on you no longer have access to model X because you're an asshole"

The commercial inference offering are also downstream of one of those 3 projects (or trt-LLM if they're nvidia). It would impact Ollama, and fireworks, together, and everyone else.

Don't tempt fate.

Re: GLM 5.2 vs. Opus

#234

Post by doe88 »

To me one shot prompting is as relevant as Strava's KOM is for cycling, i'm more interested in a good cycling performance after a 3 hours ride than a straight up 30 min record effort.

Re: GLM 5.2 vs. Opus

#235

Post by maybe_pablo »

malshe wrote: post #220
jeremyjh wrote: post #209

Earlier quoted context omitted.

No one is doing that for a model this size it would have to be so heavily quantized that it wouldn’t be useful - or you’d need to spend a half million dollars on hardware. People use hosted APIs. Open weight means cloud vendors can host it.

Can you recommend any US based cloud providers?

In HuggingChat (https://huggingface.co/chat) you can test open models for free and even test specific providers.

From there I collected the following US providers currently serving GLM 5.2:

- Together (https://www.together.ai/models)

- Fireworks (https://fireworks.ai/models)

- Featherless (https://featherless.ai/models)

Re: GLM 5.2 vs. Opus

#236

Post by scwoodal »

Der_Einzige wrote: post #231
scwoodal wrote: post #104

Earlier quoted context omitted.

> I'd rather see a coding agent that can follow steps in a plan file to a T while following guardrails and adhering to the proper coding conventions in the human reviewed spec. Guardrails/conventions should be enforced in linters, formatters, static analysis tooling; not specs/prompts.

Wrong, custom "specs" i.e. schemas, are literally all we have for "real" guardrails with LLMs. https://developers.openai.com/api/docs/guides/structured-out... Nothing else operates on the logprobs level and literally bans continuations that fail your schema.

Enforcing structured outputs from LLMs is not the same thing as using linters, formatters, static analysis to control how an agent writes code.

Re: GLM 5.2 vs. Opus

#237

Post by Der_Einzige »

scwoodal wrote: post #236
Der_Einzige wrote: post #231

Earlier quoted context omitted.

Wrong, custom "specs" i.e. schemas, are literally all we have for "real" guardrails with LLMs. https://developers.openai.com/api/docs/guides/structured-out... Nothing else operates on the logprobs level and literally bans continuations that fail your schema.

Enforcing structured outputs from LLMs is not the same thing as using linters, formatters, static analysis to control how an agent writes code.

No, it's not. It's strictly better.

Re: GLM 5.2 vs. Opus

#238

Post by metobehonest »

WarmWash wrote: post #204
twobitshifter wrote: post #145

Earlier quoted context omitted.

Taking a view from outside the USA, European companies just had Fable taken away due to US export controls, and before that Anthropic announced it is holding their data for 30 days. There is immediate value to these firms to build their infrastructure around an AI that won’t be pulled away from them. And outside of Europe, other countries are more price sensitive and don’t have the same fear of building relationships…

There is no such thing as a relationship with "chinese companies". In China there is just the State, and that is it. If the world needs any more evidence of Europe's short-sightedness, it would be them running to China to spite the US (instead of creating fertile grounds for their own tech).

No one is running to China to "spite the US". Recent geopolitical developments have shown the US to be a violent, unpredictable and unreliable partner.

Re: GLM 5.2 vs. Opus

#239

Post by post-it »

newaccountman2 wrote: post #191
post-it wrote: post #180

Earlier quoted context omitted.

The streetlight effect: > A policeman sees a drunk man searching for something under a streetlight and asks what the drunk has lost. He says he lost his keys and they both look under the streetlight together. After a few minutes the policeman asks if he is sure he lost them here, and the drunk replies, no, and that he lost them in the park. The policeman asks why he is searching here, and the drunk replies, "this is…

[flagged]

It could be a taxi driver if you like. Or an anarchist passing by on xir way to a protest.

Re: GLM 5.2 vs. Opus

#240

Post by moistoreos »

cultofmetatron wrote: post #19

I seriously dont' know all this big hullabaloo about one shot prompting. by definition, a single prompt wont' constitute the complexity of a software project. ergo, what you'll get is a series of assumptions made by the model based on preexisting code in its training corpus. I'd rather see a coding agent that can follow steps in a plan file to a T while following guardrails and adhering to the proper coding conventio…

PREACH. I have no idea why THIS has become the standard for illustrating model capabilities. It's endlessly frustrating when that was the initial objective for all these models, but, became increasingly clear over time that none of these models were ever capable of getting the desired output for complex software on the initial prompt.

The reality is: - business rules change - ideas for improvement may arise from the initial prompt - updates to submodules/functions/configs/secrets are BLOCKERS ... etc.

One shot prompting for the expecations of complete software is seemingly more and more a show of incompetence of the use of this technology. It's like trying to make my toddler eat a ham sandwich from the peanut butter & jelly I put in front of him.

Post reply on HN