Bolwin

Redacted is a bit extreme. They just haven't benched them yet

Interesting you should say this, as Microsoft just yesterday announced an update fixing this (gradual rollout).

The joy of learning something new is very much tied to said thing being meaningful or useful. You could spend the rest of your life memorizing digits of pi and you'd be spending all of it "learning something new" but you wouldn't enjoy it very much.

Kimi has sound, it's just very quiet

The original k3 also matched its respective opus

This seems more a measure of anti promp injection than general intelligence

I hope someone runs multilangbench for other models. I really want to see how open models in particular do on it. Benchmaxxing by focusing on the top 5 languages seems an easy

That's fascinating. Are these for your keyboard?

You may have gone too fast honestly. Something like parakeet runs fast enough and seems to have much lower wer

Which claude model are you comparing it to? If you're using fable max obv it's going to be slower, but like sonnet high is both faster and more intelligent

Very similar to Kimi K3 it seems, but fewer active params (80b vs 104b). I don't think they mentioned what form of attention they use

This is why I severely dislike using all pass rate. A model could be at like 93%, meeting practically every requirement except the broken ones and all pass would give it like 20%.

I use it for title generation basically. Will have to see where this one fits in

That's cause and effect, not physics.

And it's not technobabble. I like hard sci-fi. I don't pretend to understand every field of tech, nor do so my characters have the same expertise as me. Having an llm with knowledge of basically all fields is helpful.

Try fable. I haven't used it since they dropped it from the pro plan, but when I did, fable 5 casually dropped such advanced electrical and orbital mechanics knowledge in my story that I had to stop and ask it to explain

I don't really understand why Rust seems to be the go-to language for llms. Yes sure it has a good type system, but in practice llms seem to trip over it as much as I do, and given how fast llms are at writing code, that just makes compile times a much larger bottleneck.

The biggest discovery might actually be that they ignored regular caching rules and kept invalid cache suffixes and it didn't hurt performance

Look at Neuralwatt. They report energy usage with every call as well as aggregate statistics.

Why are you assuming the gpus are rented?

Anyway, they still have token based pricing if you prefer. The energy pricing is often cheaper though

This article feels like it was written by a human who has just read to much llm writing

I think it's more likely they recorded the video when the project was already done than cheated

Moonshot has already teased K3.1 so not likely

Why?

Build?

Nah these robots are for repacing the customer service jobs

Alibaba is entirely capitalistic I'm afraid

People often use this a argument against small phones like this wouldn't be amazing for any company outside apple or Samsung

What if you don't own a phone?

Another site in fact https://dontasktoask.com/

The intermediate tickers are fake but real data comes in and resets it. Its like a progress bar essentially. We don't call progress and bars fake

I don't really remember a situation, which of those models supposedly beat the other?

I still opus 4.6 though not for code

Distillation requires you to have the actual logits of each token from the teacher model, which in practice means having the model itself.

What you're describing is just synthetic data.

Note Anthropic misused the term in their post about Chinese model distillation, deliberately I assume.