1
point
ZeroCool2u
Joined 3,603 karma
FRBNY: Senior Data Scientist
https://theolinnemann.com
- ZeroCool2uLMStudio is so much better than Ollama it's silly it's not more popular.
- Imagine a Steam TV with the Steam Box simply built-in. That would be incredibly nice. The worst part of my brand new LG G5 OLED TV is the software itself. I'd pay a good deal more to have Valve responsible for the software running on my TV.
- Yeah, the reverse breakup fee is ~2.6B I believe, but the Paramount takeover doesn't have to succeed for that fee to kick in. WB just has to back out.
- Seems okay. It's no Opus 4.5 or Gemini 3 Pro according to the benchmarks. Also, still a good chance the AWS team is benchmaxing the same as last time.
Additionally, my experience with Bedrock hasn't made me a huge fan. If anything its pushed me towards OpenRouter. Way too many 500 errors when we're well below our service quotas.
- 1 point
- I've had to repeatedly tell our AWS account reps that we're not even a little interested in the Trainium or Inferentia instances unless they have a provably reliable track record of working with the standard libraries we have to use like Transformers and PyTorch.
I know they claim they work, but that's only on their happy path with their very specific AMI's and the nightmare that is the neuron SDK. You try to do any real work with them and use your own dependencies and things tend to fall apart immediately.
It was just in the past couple years that it really became worthwhile to use TPU's if you're on GCP and that's only with the huge investment on Google's part into software support. I'm not going to sink hours and hours into beta testing AWS's software just to use their chips.
- Billie is a good dog.
- Yeah, and the main problem with HEVC/H265 is the patent encumbrance. Very odd, but hopefully it's just coming a bit later.
- Interesting, I've been dealing with replacing a few on-prem HPC clusters lately. One of the things we've been looking at is OpenOnDemand. How does this compare to that? Is this primarily targeted at cluster development or can I really just make an arbitrarily large production HPC cluster with it?
- Yeah, I guess my question was a bit more nuanced. What I was curious about was if they were fully relying on normal autoscaling that any customer would get or were they manually scaling the spanner instance in anticipation of the load? I guess it's unlikely we're going to get that level of detailed info from this article though.
- But, and I'm honestly asking, you as a GKE user don't have to manage that spanner instance, right? So, you should in theory be able to just throw higher loads at it and spanner should be autoscaling?
- This is a good idea, will give it a try!
- Sure! But it's weird how far off it is in terms of capability.
- I tried the studio ghibli prompt on a photo my me and my wife in Japan and it was... not good. It looked more like a hand drawn sketch made with colored pencils, but none of the colors were correct. Everything was a weird shade of yellow/brown.
This has been an oddly difficult benchmark for Gemini's NB models. Googles images models have always been pretty bad at the studio ghibli prompt, but I'm shocked at how poorly it performs at this task still.
- Already available in the Gemini web app for me. I have the normal Pro subscription.
- Seems like their new Antigravity IDE specifically has this built in. https://antigravity.google/docs/browser
- Yes, I think one of the most important things we as consumers can do is flood the zone for companies like Netflix, Disney, and Apple and keep asking about native Steam Machine apps installed directly from the Steam store that support 4k streams.
- LM Studio will run dynamic quants from Unsloth too. Much nicer than Ollama.
- The workspace feature seems like the biggest differentiator between this and the Apple Vision Pro. Full multi window display, with what seems to be desktop app functionality? That's almost tempting.