David Graham

100% into @daisinet and there's no stopping me.

[David Graham @OriginalCoder]

Daisi LLogos now beats llama.cpp on three of six Qwen model tests using pure C#. Read the white paper.

https://t.co/PN11XCAlZy

#ai #llama #daisi #dotnet #inference #research #qwen

[daisi.ai @daisinet]

Telos is coming...

Continuous Self Learning
Creative at small and large scales
Distributed and Decentralized
Limitless context size
Can forget on demand
No pre/post training needed
Loads in under a second
10,000+ decode tokens per second
Runs on any CPU silicon
Under 50 MB of RAM

[David Graham @OriginalCoder]

@Mente_Alpha_ It's a known tactic so that the other person cant crush your hand. If you shake hands all day, and some dudes will try to crush you, it works.

[David Graham @OriginalCoder]

@TomSteyer The government's fair share is $0.

[David Graham @OriginalCoder]

@mehdirhasan No, she's saying that he shouldn't have the opportunity, nit that he doesnt have the right.

[David Graham @OriginalCoder]

@Austen Dear Strongbad... how we miss you

[David Graham @OriginalCoder]

@iselltops @mert There are people that eat healthy their entire lives, dont drink or smoke, and exercise. In two different studies, statins made no correlative reduction in heart disease and stroke.

@grok can you find the studies, please.

[daisi.ai @daisinet]

Broski is an open-source native C#, pure implementation, for a headless browser. Custom built JavaScript tokenizer and interpreter runs in a sub process and is completely sandboxed.This enables web search and fetching nearly all modern web pages from code. https://t.co/mqIunTJJ9w

[daisi.ai @daisinet]

The DAISI solutions platform is really coming together. Network is fast, secure, and stable. The product selection is growing week by week. All of our products and frameworks are free under the community licenses, or you can use our hosting services for a small fee.

[David Graham @OriginalCoder]

@joerogan sitting across a desk from @ArsenioHall talking about how having a desk was a stupid idea is a special kind of irony.

[David Graham @OriginalCoder]

@EricLDaugh Looks like they are there on their own. They should go home and stay safe. Hopefully, they do.

[David Graham @OriginalCoder]

@SuperSisi Great. I would definitely play this. Needs a minimap though or I would be lost.

[David Graham @OriginalCoder]

For the record, I don't want war with Iran.

[David Graham @OriginalCoder]

Been trying to avoid buying expensive hardware to run larger models and I finally have a way to do it. It's still slow, but it works. I can always work on efficiency later.

[daisi.ai @daisinet]

Daisi-Chain is running large models on small systems in two ways:
1. Device loads a subset of layers, then connects to the chain gang for that model. Each layer is processed locally, when available, then the chain processes the rest. Everyone gets free large model inference.

[David Graham @OriginalCoder]

While Claude is down, you might as well check out @daisinet

[daisi.ai @daisinet]

I gave everyone that has an account 250k credits to play with today. If you want some free credits, create an account on DAISI and I will get you hooked up.

[David Graham @OriginalCoder]

@theliamnissan It was all donated by a single guy that wanted to break the record so that more people have a chance to learn about Christianity. Kinda the whole point.

[David Graham @OriginalCoder]

@wonkjunkie @kylegriffin1 We're they forced to take the loan? We're they able to pay? I dont need to understand the VA loan program to use common sense. They could not afford the loans that they took. Plenty of people do it. Vets arent immune to it.

[David Graham @OriginalCoder]

@MattMMPLS @kylegriffin1 I would only complain if it impacts other people's rights.

[David Graham @OriginalCoder]

@MattMMPLS @kylegriffin1 Im not MAGA. They were able to pay it, sure, but they couldn't afford what they agreed to. Debt isnt a right. They got a loan with the backing of the US government that they would not have been fot to get otherwise. They shouldn't have taken the loan that could not afford.