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

Everything we announced at our first LlamaCon

ai.meta.com

Post reply on HN
Print view View original

21–30 of 122 posts

Re: Everything we announced at our first LlamaCon

#21

Post by mgdev »

OtherShrezzing wrote: post #11
mgdev wrote: post #10

There is a potential world where Meta uses AI as a vector to tap into the home. Like, literally building smart homes. Locally intelligent in ways that enable truly magical smart home experiences while preserving privacy and building trust. But connected in ways that facilitate pseudo-social interactions, entertainment, and commerce. Meta's biggest competitors are Apple and Amazon. This is the first clear opportunity…

>There is a potential world where Meta [is]... literally building smart homes... while preserving privacy and building trust I'm earnestly not sure what Meta are less qualified for. Building physical homes or building privacy & trust.

Visit SE Asia sometime and you'll experience a very different sentiment. Hundreds of millions of people rely on Meta to provide valuable services every day, some of them borderline essential. This is undebatable.

The outsized public hatred toward Meta is almost entirely driven by a bureaucratic, anti-technology Europe (that has finally realized that their overstepping is hurting their future) and a US political institution that needed someone to demonize to keep us all distracted.

There are very good reasons to dislike Meta and Meta products. But they're likely not the ones you're referring to.

Re: Everything we announced at our first LlamaCon

#22

Post by smcnally »

walterbell wrote: post #14
smcnally wrote: post #13

Earlier quoted context omitted.

Beyond solid benchmarks, Alibaba's power move was dropping a bunch of models available to use and run locally today. That's disruptive already and the slew of fine tunes to come will be good for all users and builders. https://huggingface.co/collections/Qwen/qwen3-67dd247413f0e2...

What's the minimum GPU/NPU hardware and memory to run Qwen3 locally?

`model.safetensors` for Qwen3-0.6B is a single 1.5GB file.

Qwen3-235B-A22B has 118 `.safetensors` files at 4GB each.

There are a bunch of models and quants between those.

Re: Everything we announced at our first LlamaCon

#24

Post by Keyframe »

throwaw12 wrote: post #4

Feels like Meta is going to Cloud services business but in AI domain. They resisted entering cloud business for so long, with the success of AWS/Azure/GCP I think they are realizing they can't keep at the top only with social networks without owning a platform (hardware, cloud)

If Lidl can venture into cloud business, I guess so can Meta.

Re: Everything we announced at our first LlamaCon

#25

Post by philipkglass »

ahmedfromtunis wrote: post #23

Does anyone use llama as their primary model for any usecase? Maybe it's my fault for not spending much time with it, but I still couldn't find the applications for which llama has an advantage over the competition.

I recently needed to classify thousands of documents according to some custom criteria. I wanted to use LLM classification from these thousands of documents to train a faster, smaller BERT (well, ModernBERT) classifier to use across millions of documents.

For my task, Llama 3.3 was still the best local model I could run. I tried newer ones (Phi4, Gemma3, Mistral Small) but they produced much worse results. Some larger local models are probably better if you have the hardware for them, but I only have a single 4090 GPU and 128 GB of system RAM.

Re: Everything we announced at our first LlamaCon

#26

Post by marcodena »

ahmedfromtunis wrote: post #23

Does anyone use llama as their primary model for any usecase? Maybe it's my fault for not spending much time with it, but I still couldn't find the applications for which llama has an advantage over the competition.

Here https://research.atspotify.com/2024/12/contextualized-recomm... ;)

Re: Everything we announced at our first LlamaCon

#27

Post by mvieira38 »

ahmedfromtunis wrote: post #23

Does anyone use llama as their primary model for any usecase? Maybe it's my fault for not spending much time with it, but I still couldn't find the applications for which llama has an advantage over the competition.

It's pretty popular in the local LLM space

Re: Everything we announced at our first LlamaCon

#28

Post by galeos »

philipkglass wrote: post #25
ahmedfromtunis wrote: post #23

Does anyone use llama as their primary model for any usecase? Maybe it's my fault for not spending much time with it, but I still couldn't find the applications for which llama has an advantage over the competition.

I recently needed to classify thousands of documents according to some custom criteria. I wanted to use LLM classification from these thousands of documents to train a faster, smaller BERT (well, ModernBERT) classifier to use across millions of documents. For my task, Llama 3.3 was still the best local model I could run. I tried newer ones (Phi4, Gemma3, Mistral Small) but they produced much worse results. Some large…

How did you find ModernBERT performance Vs prior BERT models?

Re: Everything we announced at our first LlamaCon

#29

Post by philipkglass »

galeos wrote: post #28
philipkglass wrote: post #25

Earlier quoted context omitted.

I recently needed to classify thousands of documents according to some custom criteria. I wanted to use LLM classification from these thousands of documents to train a faster, smaller BERT (well, ModernBERT) classifier to use across millions of documents. For my task, Llama 3.3 was still the best local model I could run. I tried newer ones (Phi4, Gemma3, Mistral Small) but they produced much worse results. Some large…

How did you find ModernBERT performance Vs prior BERT models?

I didn't try original BERT at all because I didn't get good results from any LLMs on small document excerpts, so I assumed that a substantial context was necessary for good results. Traditional BERT only accepts up to 512 tokens, while ModernBERT goes up to 8192. I ended up using a 2048 token limit.

Re: Everything we announced at our first LlamaCon

#30

Post by littlestymaar »

mvieira38 wrote: post #27
ahmedfromtunis wrote: post #23

Does anyone use llama as their primary model for any usecase? Maybe it's my fault for not spending much time with it, but I still couldn't find the applications for which llama has an advantage over the competition.

It's pretty popular in the local LLM space

It used to, but Llama 4 is useless for local LLM for most people.
Post reply on HN