1171
Speculative KV coding: losslessly compressing KV cache by up to ~4×
Lossless compression of a target model's KV cache by up to 4×, using a cheaper predictor model to drive an arithmetic coder.
Top posts
Every blog post that took off on Hacker News, ranked by the votes of its best submission. 1,521 posts.
1172
The efficient frontier of LLM inference
Inference techniques either move a deployment along the latency–throughput frontier or push the entire frontier out, creating more efficiency to allocate.
1173
What Meta got right with Muse
Design matters more than ever in consumer AI, as long as you have money to play.
1174
1175
1176
Claude for Creative Work
New connectors let Claude work inside Adobe Creative Cloud, Blender, Autodesk Fusion, SketchUp, and music production tools to support creative workflows.
1177
Rethinking legal education in the AI era
OverviewArtificial Intelligence is already impacting higher education and the legal profession, and the pace of change appears only to be accelerating. It is thus critical for us to use this moment to carefully reflect on how legal education should adapt. This memo describes the approach we are taking to legal…
1178
Cord: Coordinating Trees of AI Agents
AI agents are good at doing one thing at a time. Give Claude a focused task and it performs. But real work isn’t one task. It’s a tree of tasks with dependencies, parallelism, and context that needs to flow between them.
1179
Letting Claude play text adventures
Experiments in cognitive architecture.
1180
Advancing finance with Claude Opus 4.6
With Claude Opus 4.6, finance teams get better reasoning on complex analyses, cleaner first-pass deliverables, and new tools built for where analysts actually spend their time.
1181
The safest job from AI may be writing
Today, tech folk are scrambling to change their workflows to meet newly inflated 5X productivity quotas, while getting pummeled under the co...
1182
People Hate AI Art
If your initial reaction to reading that and seeing that is some variation of "ughhh" or rolling your eyes or "fuck this guy" congrats. You are normal.
1183
Locating a Photo of a Vehicle in 30 Seconds with GeoSpy
This is just a short description for your post which you can edit depending on your needs via the Framer CMS.
1184
Dell's version of the DGX Spark fixes pain points
Dell sent me two of their GB10 mini workstations to test: In this blog post, I'll cover the base system, just one of the two nodes. Cluster testing is ongoing, and I'll cover things like AI model training and networking more in depth next year, likely with comparisons to the Framework Desktop cluster and Mac Studio…
1185
The Future of AI
The Parents’ Paradox: AI, Ethics, and the Limits of Machine Morality This post is based on a talk I gave at The AI & Automation Conference in London on February 25, 2026, and my slides. A…
1186
Everything is an AI scam now
Whenever you watch a video on YouTube now, you will likely get an advert for a scammy product. In the past few weeks I've seen:
1187
U of T researchers demonstrate AI worm could target any online device
A team of researchers at the University of Toronto has discovered a new class of cyberthreat that gives hackers more power and reach at far less cost. It can be built with free AI models. Every online device is a potential target. And current cyber defences are not yet ready for it.
1188
LLMs are not superhuman at reading and writing neural network weights
A recent paper argues that frontier LLMs can work with neural network weights directly. This seems superhuman, but it isn’t. The paper shows that GPT-6 Astra…
1189
Measuring progress toward AGI: A cognitive framework
Google DeepMind proposes a cognitive framework to evaluate AGI and launches a Kaggle hackathon to build capability benchmarks
1190
Run NanoClaw in Docker Sandboxes
Every agent gets its own isolated container inside a micro VM. No dedicated hardware needed. No complex setup.
1191
Anthropic installed a spyware bridge on my machine?
Anthropic's Claude Desktop silently installs a Native Messaging bridge into seven Chromium browsers, including browsers Anthropic's own documentation says it does not support, and browsers the user has not even installed.
1192
AI-ready biological data: $1.8B global commitment
Biohub, DOE, NIH, and partners will generate open, standardized data to train AI models that predict how cells respond to interventions.
1193
Inside vLLM: Anatomy of a High-Throughput LLM Inference System (2025)
From paged attention, continuous batching, prefix caching, specdec, etc. to multi-GPU, multi-node dynamic serving at scale.
1194
A Misalignment of AI in Mathematics
I am proud to be among the list of 25 initial signatories — all Fields Medallists — to the declaration below, which grew out of discussions between ourselves over the last week. We have…
1195
Vibecoding #2
I feel like I got substantial value out of Claude today, and want to document it. I am at the tail end of AI adoption, so I don't expect to say anything particularly useful or novel. However, I am constantly complaining about the lack of boring AI posts, so it's only proper if I write one.
1196
1197
Strands Harness
Strands harness is a fully assembled, customizable, state-of-the-art agent you run locally or deploy anywhere.
1198
We're training a Figma-fluent editing model
Here, we’re sharing the research behind how we’re translating that to design editing workflows.
1199
Tech valuations are back to pre-AI boom levels
The chart below compares the forward P/E ratios for the S&P 500 and the S&P 500 Information Technology sector. Subscribe for daily updates.
1200
Step 5 Preview, a 1M-context MoE from StepFun, shows up on OpenRouter
Step 5 Preview is StepFun's flagship model for agentic work, built on a sparse Mixture-of-Experts architecture (27B active / 600B total parameters).