Context Windows

Feeds to Scour
SubscribedAll
Scoured 121 posts in 10.8 ms

KaiFelixBennett/gemma4-turboquant-rdna4: Run Gemma-4-31B at full 256K context on a $1,400 AMD RDNA4 GPU (gfx1201): TurboQuant KV cache + HIP-graph-safe Flash-Attention for llama.cpp, fully measured on real hardware.

 🤖LLM  Content type: Code
github.com··Hacker News

How we fight GPU scarcity without compromise

 🤖LLM  Content type: Blog
equixly.com··Hacker News

FlashMemory-DeepSeek-V4: Lightning Index Ultra-Long Context via Lookahead Sparse Attention

 🤖LLM  Content type: Academic
arxiv.org··Hacker News

Larger context windows and configurable reasoning levels for GitHub Copilot - GitHub Changelog

 💬LLMs  Content type: Blog
github.blog··Hacker News
Less-relevant results

Tokenminning: Because Tokenmaxxing Is a Bad Idea

 ✍️Prompt Engineering

Inferoa AI harness claimed 90% cache savings. We ran it and measured 97.8%

 🤖LLM
zozo123.github.io··Hacker News

manavgup/context-analyzer: Context window usage analyzer for Claude Code — MCP server + interactive dashboard

 🔌MCP  Content type: Code
github.com··Hacker News

How LLMs Actually Work: A Friendly Map for Humans • oreoro

 💬LLMs

LLM are universal simulators

 🤖LLM

Claude Fable 5 “Feels Next Level” via @sejournal, @martinibuster

 🎭Anthropic Claude

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

 🤖LLM

A system programmer’s guide to LLM inference

 🤖LLM  Content type: Blog

LLM Research Papers: The 2026 List (January to May)

 🤖LLM  Content type: News

princezuda/-RequiemGPT-: Fully open source and open weights built and trained by fable five with one prompt. An experience in how AI actually works

 📄Plain Text  Content type: Code
github.com··Hacker News

The economics of speculative decoding

 🤖LLM  Content type: Blog

Claude Fable 5 Launches at #1 on the Artificial Analysis Intelligence Index

 🎭Anthropic Claude  Content type: News

Context Sculpting

 🎭Anthropic Claude  Content type: Blog

Ask HN: Is it feasible to run a model on device for complete privacy?

 💬LLMs  Content type: Discussion

linzhiqiu/t2v_metrics: Evaluating text-to-image/video/3D models with VQAScore

 🤖LLM  Content type: Code
github.com··Hacker News

Tangram: Unlocking Non-Uniform KV Cache for Efficient Multi-turn LLM Serving

 🤖LLM  Content type: Academic
arxiv.org··Hacker News

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