AI

Feeds to Scour
SubscribedAll
Scoured 91 posts in 11.1 ms

Pythia 1.4B reproduces 3.6% of training samples verbatim given 950-token prompts

 🔓Open Source  Content type: Blog
ret2libc.com··Hacker News

a desktop GUI to browse, search, & visualize your vector databases

 🗃Databases

The Cosmological Hart-Tipler Conjecture

 🔍AI Interpretability  Content type: Academic
arxiv.org··Hacker News

I open-sourced my UFC prediction model, code, and database after 5 years of work

 LLM Optimization
mcinerney.ai··Hacker News

GPT-2: Too Dangerous To Release (2019)

 LLM Optimization  Content type: Blog

Introducing Piper: A Programmable Distributed Training System

 Model Efficiency  Content type: Academic  Content type: Blog

China's Xiaomi MiMo Is Now 15X Faster Than ChatGPT and Claude (4 minute read)

 Model Efficiency  Content type: News
decrypt.co··Hacker News

Running Qwen 35B MoE at 450k Context on a Single 32GB GPU

 LLM Optimization

"Trust Us" Is Not a Control Surface: Anthropic and the Case for Open Weights

 🗃Databases

Introducing North Mini Code: Cohere’s First Model For Developers

 ✍️Prompt Engineering  Content type: Blog

GGUF vs GPTQ vs AWQ: The Plain-English Guide to LLM Quantization (and Which One to Pick)

 LLM Optimization

Researchers trained an open source AI search agent, Harness-1, that outperforms GPT-5.4 on recalling relevant information

 🗃Databases

If Claude Fable stops helping you, you'll never know

 ✍️Prompt Engineering  Content type: Blog

Tejas-TA/predikit: The missing bridge between your ML models and your AI agents.

 🐍Python  Content type: Code
github.com··Hacker News

Fine-tuning vs RAG vs MeMo: Where should LLM Knowledge Live?

 ✍️Prompt Engineering
pub.towardsai.net
·

Claude Fable 5 and new AI safety fables

 LLM Optimization  Content type: News

"North Mini Code"; open weights, 30B param, Canadian coding model

 Model Efficiency  Content type: Blog

Gemma 4 QAT models: Optimizing model compression for mobile and laptop efficiency

 LLM Optimization  Content type: News  Content type: Blog
blog.google··Hacker News

Claude Fable 5 silently degrades its own performance on frontier AI work

 Model Efficiency  Content type: News  Content type: Blog

Stack Overflow didn't just help AI learn to code

 ✍️Prompt Engineering

Keyboard Shortcuts

Navigation

Next / previous item
j/k
Open post
oorEnter
Preview post
v

Post Actions

Love post
a
Like post
l
Dislike post
d
Undo reaction
u
Save / unsave
s

Recommendations

Add interest / feed
Enter
Not interested
x

Go to

Home
gh
Interests
gi
Feeds
gf
Likes
gl
History
gy
Changelog
gc
Settings
gs
Browse
gb
Search
/

General

Show this help
?
Submit feedback
!
Close modal / unfocus
Esc

Press ? anytime to show this help