Exploring FlashAttention-3/4 optimizations on RTX GPUs
I was curious whether any of the FA-3/4 optimizations transfer to RTX GPUs. vLLM/SGLang attention falls back to FA-2 on consumer cards (FA-3 and FA-4 are…
Latest
NVIDIA Puzzle-75B-A9B NVFP4 at 132 t/s on 3×3090 — Why is this size category a desert otherwise?
TLDR: 75B-total / 9B-active MoE is the perfect shape for multi-24GB rigs, and almost nobody ships it. Qwen 27B is a great model and punches way…
RT Rayan @ICML: Meta made a “minor” release to Muse Spark, there’s nothing minor about it. Lots to parse here: – This model is so fucking cheap I a…
RT Rayan @ICMLMeta made a “minor” release to Muse Spark, there’s nothing minor about it. Lots to parse here:- This model is so fucking cheap I…
OpenMOSS-Team/MOSS-Transcribe-Diarize · Hugging Face
MOSS-Transcribe-Diarize 0.9B is an end-to-end audio understanding model for long-form multi-speaker transcription, diarization, timestamps, and acoustic event awareness. Given an audio or video file, the model…
AI content is everywhere on social media, especially LinkedIn
Article URL: https://www.pangram.com/blog/ai-in-your-feed Comments URL: https://news.ycombinator.com/item?id=48847940 Points: 84 # Comments: 62
Show HN: Reverse-engineering web apps into agent tools
Hey HN! We built a browser-based agent that runs inside an authenticated web app, watches how the app calls its own APIs, and automatically turns those…
AWS GraphRAG deployment cuts drug research cycles by 87%
A recent AWS GraphRAG deployment reduced drug research and development cycles in pharmaceutical environments by 87 percent. This acceleration is achieved by integrating previously separated proprietary…
Some Thoughts on The Environment Problem in Agent Training
As Large Language Models move away from being chat interfaces and become increasingly autonomous actors in the real world, a few insights about evaluation and training…
The Cube Theory of Partially Grasped Concepts
[I was aiming for legibility to a limited extent only. This post got extracted from a bigger post I've been writing and is meant mostly as…
RT Emma Scharfman: I'm excited to share that I'm joining @huggingface as a ML Research Engineer on the science team 🚀🤗 My goal is to bridge the …
RT Emma ScharfmanI'm excited to share that I'm joining @huggingface as a ML Research Engineer on the science team 🚀🤗My goal is to bridge the gap…
RT Nomadic AI: Fine-grained subtask annotations improve VLA performance. We trained a vision + proprioception model that’s now SOTA at generating the…
RT Nomadic AIFine-grained subtask annotations improve VLA performance.We trained a vision + proprioception model that’s now SOTA at generating them, generalizing across embodiments.93.1 F1@50 on REASSEMBLE,…
Debate with Self-Play Best-of-N Optimization
Debate is a proposed protocol for scalable oversight. As tasks outrun direct supervision, labs are increasingly likely to train against protocols like it. Our concern is…
Launch HN: Context.dev (YC S26) – API to get structured data from any website
Hi Hacker News, I’m Yahia.I built Context.dev (https://www.context.dev/) to make it really easy to integrate web data into your products and agents.Here’s a demo video: https://www.tella.tv/video/build-faster-with-context-dev-api...Since…
RT Maithra Raghu: Excited to be releasing FrontierFinance, the largest and most challenging open benchmark for evaluating AI agents across the full in…
RT Maithra RaghuExcited to be releasing FrontierFinance, the largest and most challenging open benchmark for evaluating AI agents across the full investment workflow!FrontierFinance is substantially harder…
Hy3
Article URL: https://hy.tencent.com/research/hy3 Comments URL: https://news.ycombinator.com/item?id=48847552 Points: 11 # Comments: 1
How to Write an Email
Article URL: https://blog.dannycastonguay.com/how-to-write-an-email/ Comments URL: https://news.ycombinator.com/item?id=48847536 Points: 3 # Comments: 1
Show HN: Policy enforcement for Claude Code, Cursor, and Codex
Show HN: Runtime authorization for Claude Code, Cursor, and CodexHi HN, Fernando and I built Kastra. Kastra intercepts AI agent tool calls and evaluates them against…
A Possible Future for Damn Interesting
Article URL: https://www.damninteresting.com/a-possible-future/ Comments URL: https://news.ycombinator.com/item?id=48847511 Points: 3 # Comments: 0
RT Katie Miller: Mark Zuckerberg hasn’t posted since 2023. X is the global town square — this is where news happens.
RT Katie MillerMark Zuckerberg hasn’t posted since 2023. X is the global town square — this is where news happens.Mark Zuckerberg: (1) Today we're releasing Muse…
Opinionated and Easy Pi.dev Configuration
Article URL: https://lazypi.org/ Comments URL: https://news.ycombinator.com/item?id=48847407 Points: 9 # Comments: 0
Coordination Without Consolidation: On Systems of States [pdf]
Article URL: https://isonomiaquarterly.com/wp-content/uploads/2026/06/iq-4.2-summer-2026-macdonald-coordination-without-consolidation.pdf Comments URL: https://news.ycombinator.com/item?id=48847405 Points: 6 # Comments: 1
"OpenAI has arrived at a new solution that seems to combine the best parts of the two solutions that human top performers ultimately reached."
"OpenAI has arrived at a new solution that seems to combine the best parts of the two solutions that human top performers ultimately reached."Yoichi Iwata: AWTF…
Stop Using reCAPTCHA – It's Not a Real Captcha
Article URL: https://na0341.bearblog.dev/info/stop-using-recaptcha/ Comments URL: https://news.ycombinator.com/item?id=48847368 Points: 4 # Comments: 4
will there be a day for the COMMODITY LORDS next?
will there be a day for the COMMODITY LORDS next?Joe Weisenthal: Iconic chart tbh