- cross-posted to:
- hackernews@lemmy.bestiver.se
- cross-posted to:
- hackernews@lemmy.bestiver.se
You must log in or register to comment.
Whose premises? Release the GPUs and make them cheap again you motherfuckers
“Where are the engineers” to manage this on prem hardware?
Well, you laid them off, you fuck.
They’re going to use AI to install the AI.
Recursive Self Installation
IBMs a bit weird. They fired a shitload of them a while back.
Then they said they’re only replacing junior roles.
Then I think they’re hiring a lot more juniors because they need to build the next seniors.
I didn’t really follow too closely but that last statement is at least something I can respect.
[edit] I think mushroommunk is absolutely on point [/edit]
It’s not really something you should respect. It’s PR shell games. They fired senior staff to replace them with cheaper juniors. Not because they actually think they’re training the future but because they think they can get the same work out of them+AI
I mean, obviously, but I guess the only reason they’re saying this is because their own models aren’t that great (IBM granite sucks) and this way they can still sell consulting services.
Maybe I’m just sleep deprived, cynical, or thinking too deep into it, but I don’t even like how they are using the words “on-prem” throughout the article instead of simply saying they’re running local LLMs. Seems like such a weird choice of wording that is meant to subtly steer less informed people away from looking into alternatives to the subscription based LLM services, as most casual people wont associate it as the same thing and ignore it and move on. Even if that’s the correct terminology for that specific field, I feel like most people would just naturally say “Yes, we are running the LLM servers locally here” as that’s the only way I’ve ever heard people say it even online.
For a company “on-prem” might not mean the same as “local”. You might run them in your company data center or in a colocated server (rented space in a data center). In either of those cases, the data center might be physically close to you, but it could be on a different continent.
“On-prem” implies that the company has physical access/control of the server.
Ah okay thank you for clearing that up for me. Makes me feel a bit less crazy now haha it was a bit off-putting with the amount of times “on-prem” was used here!
“On-prem” is the Enterprise IT antonym term of “cloud”, as in public cloud, aka hyperscalers (AWS, Azure, GCP, OCI).
“On-premises” as it refers to enterprise hw/sw, has been a phrase much longer than “local LLM”
Even though everyone moved to cloud, most companies keep some portion of their infrastructure on-prem, like prod env in the cloud and on-prem test env/backups, or vice versa
Maybe most consumers wouldn’t recognize the phrase, but industry professionals certainly would
Of course IBM has a hardware solution for you
Krishna also predicted demand for on-prem AI will spur sales of Big Blue’s Spyre mainframe accelerator cards.
Yep. This is why Nvidia is preaching open weights literally every chance they get, too.
Still. It’s a better stance, even if for selfish reasons.
Nvidia is the shovel maker. They don’t care who buys the shovels.
Guy selling fruit says fruit is juicy more breaking news at 11
“It’s a simple question for any enterprise. Would you like something that’s 10 percent of the cost if it’s 90 percent as good as the best, or do you want to pay that 10 times higher cost?”
That really depends…
90% is disasterous. It may sound good because it’s “just” 10% off 100%, but it’s awful. Two iterations reduces that to ~0.01%< but that’s still pretty bad. Obviously the more complex, the higher the risk and the more demand and time per iteration. The machine also has no concept of criticality, so how much is being saved with three or even four iterations to assure failure is basically zero?
“Then the question is: how many tasks require that? I would tell you some do, but the vast majority, summarizing an email, examining a contract, answering a customer call, composing an email, all of those can be done as well with the much cheaper answer.”
Imagine 10,000 emails an hour through your enterprise and 1,000 times It fucks up. Anything from missing info to misinterpreting to just straight up modifying/deleting. No concept of criticality per email. 8,000 in a 9–5 office. 24,000 a day for a global enterprise. That’s what 90% looks like.
Then you got “examining contracts” and all the other stuff he mentioned…
Don’t assume the expensive model is 100% accurate. The cheap model is 90% as accurate as the expensive model, which is already less than 100% accurate.
It doesn’t literally mean “90% failure rate vs 100%”
In the HLE benchmark, for example, we may be talking about 45% test score for MiMo Pro or GLM or something, vs 55% for Sonnet and 61% for maxed out Claude.
But further complicating that, those particular open weight models do better than Claude in one particular hallucination benchmark, meaning Claude/ChatGPT might be worse at RAG/extraction type workloads. It’s like this across the board, with different models trading blows in different areas.
By “90%,” he really means “it’s so close the better model depends on the particular application, workload, harness, sampling configuration” and such, and that none of these models are close to “100% accuracy” across the board.
One other thing: Anthropic, OpenAI, and Grok have all been accused of “dumbing down” models under high demand to speed up inference. If your application depends on them, and they are the only provider, that could throw a wrench in the workflow.
But open weights? Pick your provider. You get an API as consistent as you are willing to pay for.
In other words, that “100%” isn’t really representing the potential inconsistency one gets being locked to a single provider.
