HackerTrans
TopNewTrendsCommentsPastAskShowJobs

moondistance

no profile record

Submissions

VibeThinker-3B achieves 80.2 on LCBv6

twitter.com
8 points·by moondistance·28 gün önce·3 comments

Qwen 3.5: small models with impressive performance

twitter.com
6 points·by moondistance·4 ay önce·0 comments

Language Models Are Injective and Hence Invertible

arxiv.org
1 points·by moondistance·9 ay önce·2 comments

Autonomous Trash Can

twitter.com
20 points·by moondistance·12 ay önce·1 comments

Grok 4 Benchmarks

twitter.com
3 points·by moondistance·geçen yıl·0 comments

Jensen Huang – Nvidia GTC 2025 Keynote

nvidia.com
75 points·by moondistance·geçen yıl·74 comments

Exaone Deep 32B – beats DeepSeek 671B

lgresearch.ai
3 points·by moondistance·geçen yıl·1 comments

DeepSeek-R1 at 3,872 tokens / second on a single Nvidia HGX H200

blogs.nvidia.com
13 points·by moondistance·geçen yıl·1 comments

ByteDance Doubao-1.5-pro matches GPT 4o benchmarks at 50x cheaper

twitter.com
4 points·by moondistance·geçen yıl·0 comments

US-China Commission top recommendation: Manhattan project for race to AGI [pdf]

uscc.gov
4 points·by moondistance·2 yıl önce·0 comments

AI scans RNA 'dark matter' and uncovers 70k new viruses

nature.com
1 points·by moondistance·2 yıl önce·0 comments

Llama 405B 506 tokens/second on an H200

developer.nvidia.com
21 points·by moondistance·2 yıl önce·5 comments

SenseNova 5.5 claims SOTA LLM benchmark results

twitter.com
2 points·by moondistance·2 yıl önce·0 comments

New AI Training Technique Is Drastically Faster, Says Google

decrypt.co
84 points·by moondistance·2 yıl önce·38 comments

Nvidia open source LLM Nemotron 4 340B at top of the charts [pdf]

d1qx31qr3h6wln.cloudfront.net
17 points·by moondistance·2 yıl önce·1 comments

Uncensor Any LLM with Abliteration

huggingface.co
4 points·by moondistance·2 yıl önce·0 comments

comments

moondistance
·28 gün önce·discuss
Paper: https://huggingface.co/papers/2606.16140
moondistance
·11 ay önce·discuss
How does this compare with Exowatt? Assume fresnel lenses are more efficient.
moondistance
·geçen yıl·discuss
Chris Crawford is also famous for the Dragon Speech :) https://youtu.be/CBrj4S24074?si=Ph12RpW8BKsh8-qS
moondistance
·geçen yıl·discuss
Yes, big time, but there continues to be lots of progress.

Most importantly, models are maturing, and this means less custom optimization is required.
moondistance
·geçen yıl·discuss
(this is also why AMD popped 10% at open yesterday - this is a new development and talks from their 2025 "Advancing AI" event were published late last week + over the weekend)
moondistance
·geçen yıl·discuss
Yes, for many applications.

Meta, OpenAI, Crusoe, and xAI recently announced large purchases of MI300 chips for inference.

MI400, which will be available next year, also looks to be at least on par with Nvidia's roadmap.
moondistance
·geçen yıl·discuss
This is from 2020.
moondistance
·geçen yıl·discuss
Thanks :)
moondistance
·geçen yıl·discuss
Where do you find/buy land? How do you vet purchases? Can you point to a few websites, etc.? Thanks!
moondistance
·2 yıl önce·discuss
Happy New Year!

I know so few of you but feel like you’re my best friends. Love to all.
moondistance
·2 yıl önce·discuss
Microsoft is negotiating equity in OpenAI as part of the switch to a for-profit. Non-zero chance this is a negotiation flex.
moondistance
·2 yıl önce·discuss
Significant further optimizations. FP8!
moondistance
·2 yıl önce·discuss
VP Research Barret Zoph and Chief Research Officer Bob McGrew also announced their departures this evening.
moondistance
·2 yıl önce·discuss
I'd like to build a chatbot arena with pro-humanity questions.
moondistance
·2 yıl önce·discuss
This is the letter: https://righttowarn.ai/
moondistance
·3 yıl önce·discuss
Years of thinking on it, observing tech progress, and speaking with lots of people. Plus the Ship of Theseus thought experiment.
moondistance
·3 yıl önce·discuss
We will need to upload ourselves and/or have robots grow us when we get to our destination. We will get there. It sounds crazy until it doesn’t.