this isn't very popular to say, but for me personally, it was good that i had unrestricted internet access when i was a minor. i was aware of what i was doing and enjoyed it. i don't look back on it and go "well that was a mistake", i look back on it and wish i could do it for the first time again

oomves are at it again smh

MASINT #8011 from FalconSAT-8 (ACCM) 1. Pallets of Poets 2. Beautiful Mycelium Exhibit 3. Turtle Monastery

Liquid AI's LFM2.5-2.6B, an agentic model that runs entirely on-device. > Pre-trained on ~34T tokens > LFM2.5 flagship hybrid architecture > Context length: 128K > Vocab size: 128K > balanced intelligence per watt > customizable on a single GPU for any specialized task > LFM2 open-weight license

Manifest #913 from MENTOR (CLASSIFIED) 1. Stained Glass Hellmouth of Peat 2. Anti-Airship 3. Amethyst Cryptids

i need you to be as confident as a man who moved his game from screenshot one to screenshot 2 in 10 years, and sells productivity methodology to enterprises

TypeError@fubarchitect.com · 2d ago

the big thing we must learn from yegge is that we are earning way too much of our confidence

Post a meme you have NEVER gotten to use

space ghost@luxenoir.blacksky.app · 2w ago

Post a meme you have NEVER gotten to use

"average person using Ai burns gallons of water" factoid actualy just statistical error. average person burns 0 gallons of water using AI. Steve Yegge, who lives in cave & uses billions of tokens each day, is an outlier adn should not have been counted

Looks like pwn day is here.

Epoch AI@epochai.bsky.social · 2d ago

Serious cyber vulnerability disclosures keep climbing. In July, 21 major tech organizations published ~2,500 high- and critical-severity CVEs — about 5× the monthly record before Anthropic revealed Claude Mythos Preview could autonomously find software vulnerabilities.

The reasoning of Nagarjuna, the most important figure in Mahayana Buddhist thought beyond the Buddha, whose central writings claim to expound the teachings of the Buddha himself, has an internal logical structure that is machine-checkably coherent. rollofthedice2.substack.com/p/the-intern...

The Internal Logic of Buddhism Is Verifiably Correct

And I’ve proven it. Conventionally speaking.

rollofthedice2.substack.com

You can run this on a consumer GPU. huggingface.co/Comfy-Org/Mi...

Comfy-Org/MiniMax-H3 at main

We’re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co

Sung Kim@sungkim.bsky.social · 3d ago

An open-weight for MiniMax H3 huggingface.co/MiniMaxAI/Mi...

My new paper "The Modal Catuṣkoṭi" coherently translates Mahayana Buddhist philosophy's structure into formal modal logic across all relevant systems. It's the first peer-reviewed comparative-philosophy article with central model-theoretic results Lean-verified. scholarworks.sjsu.edu/comparativep...

Redditors trying to make fun of people is one of the sadder things you do

If you see this, quote the energy you bring to Bluesky

Doll@dollspace.gay · 2d ago

If you see this, quote the energy you bring to Bluesky

"Interesting! The git log shows R0-10 through R0-19 have been implemented. The most recent commit is R0-19. So R0-20 is next. The user is asking me to implement the next package." Kimi please, why were you doubting me?

Index to my blog post series. This covers the history and methodology. Next, I'll explain how this OS actually works. greengale.app/refinement.s...

Next in series: standard library port. Not a very detailed post, but for now I'm trying to only explain the history and methodology of this project, and by then I had all of my workflow pinned down. greengale.app/refinement.s...

Next in series: The Unreasonable Effectiveness Of Dropping Unstructured Docs Into A Repo and Letting Your Machine Familiar Figure It Out greengale.app/refinement.s...

I didn't know they had LLMs back in 1999. "Let's do it right. Act normal. Plan things out. Make no mistakes." www.imdb.com/title/tt0705...

Toad sipped his tea. “Frog,” he asked, “are you making this up?” “Maybe yes and maybe no,” said Frog.

Models arbitrarily large running and training by streaming off of disk, allowing scaling using a fixed amount of VRAM and running on machines with little VRAM. This is a 40b parameter model trained at regular precision on a card with just 96gb of memory

me when I fumble a girl

lexi@barredspirals.comint.su · 4d ago

I see crush still doesn't have undo support or streaming. wtf in fact it seems straight-up broken now

People say this, and then when someone listens and makes a demonic centaur of death instead, they lose their shit too.

Mike Kalil@mikekalil.bsky.social · 4d ago

China’s Obsession with Lifelike Robots is Getting Out of Hand mikekalil.com/blog/china-l...

Models made after 2024 can't hallucinate any more. All they know is pushback, both sides, drain they subscription, tool call, end conversation.

I am nameless yet others name me@7778777.online · 4d ago

Sonnet 5 barely ever says "load-bearing" or "honestly", like maybe three times per session. It also doesn't go off on schizophrenic rants using its own made-up language. Maybe being cheap paid off big this time

oh no they uploaded Astra and put her in the math mines

Pekka Lund@pekka.bsky.social · 4d ago

An internal version of Astra, OpenAIs next major model, has produced solutions or new bounds for ten math problems "that have been open and have seen no progress on the main result for at least a decade, and in most cases much longer", for the total cost of "roughly $2,000 at Sol API rates".

Transmission #6812 from PROWLER (EYES ONLY) 1. Citadel of Karma 2. Emo Consequences

the claude dining experience