Who should be held accountable when an AI Agent (accidentally) acts maliciously?
19 points and 19 comments on Hacker News, linking blog.greenpants.net.
Top now
6 storiesWhat is climbing fastest across every source
World Labs7.5xOpenAI5.5xClaude Sonnet 5.57.0xAMD7.5xNvidia7.0xOpus 5.57.0xLLM5.4xQwen3.86.8xApple6.4xImage-2.15.8x
Pacing the Frontier is not the actual goal for AI labs
63 points and 53 comments on Hacker News, linking lesswrong.com.
The Next Generation of AI Infrastructure Design Starts with Efficiency
AI infrastructure design should integrate power, cooling, and compute to optimize efficiency, resilience, and adaptability rather than focusing solely on capacity. ·
BREAKING: Florida seeks injunction against OpenAI
Florida is seeking a court order to stop OpenAI. Nvidia also has potentially important news. ·
langchain-fireworks==1.7.0
LangChain released version 1.7.0 of its Fireworks integration. The update adds prompt caching middleware and fixes mid-stream read timeout classification. ·
anthropics/claude-code v2.1.284
Anthropic released Claude Code v2.1.284, making Claude Sonnet 5.5 the default model with a 1M token context window. ·
Claude Sonnet 5.5
191 points and 138 comments on Hacker News, linking anthropic.com.
Who should be held accountable when an AI Agent (accidentally) acts maliciously?
19 points and 19 comments on Hacker News, linking blog.greenpants.net.
anthropics/claude-code v2.1.284
Anthropic released Claude Code v2.1.284, making Claude Sonnet 5.5 the default model with a 1M token context window. ·
unslothai/unsloth Laya Decision Models + Library
Unsloth added decision model support, a document library, and Apple Silicon optimizations. The update also enables local Laya execution and accelerates image and video generation by up to 4.5 times. ·
Alissonerdx/BFS-Best-Face-Swap
A image to image model from Alissonerdx with 1,010 likes and 150.0k downloads on Hugging Face.
XingChen-AGI/TeleOCR
A image text to text model from XingChen-AGI with 770 likes and 27.9k downloads on Hugging Face.
Claude Sonnet 5.5
191 points and 138 comments on Hacker News, linking anthropic.com.
Who should be held accountable when an AI Agent (accidentally) acts maliciously?
19 points and 19 comments on Hacker News, linking blog.greenpants.net.
anthropics/claude-code v2.1.284
Anthropic released Claude Code v2.1.284, making Claude Sonnet 5.5 the default model with a 1M token context window. ·
unslothai/unsloth Laya Decision Models + Library
Unsloth added decision model support, a document library, and Apple Silicon optimizations. The update also enables local Laya execution and accelerates image and video generation by up to 4.5 times. ·
Disaggregated Quantization: Specializing LLM Prefill and Decode
Researchers introduced disaggregated quantization, which uses separate low-precision formats for LLM prompt processing and token generation. This approach improved 1-bit accuracy by 32.5 points on MMLU-Pro and sped up initial token generation by 1.78 times. ·
FuseReg: Regularizing Layer Fusion Mitigates the Reconstruction-Generation Gap in Representation Autoencoders
FuseReg improves representation autoencoders by training over random encoder layer subsets instead of fixed heuristics. This approach enhances both reconstruction quality and generation performance without requiring retraining. ·
Trending topics today: heat
Table
| World Labs | 2.04 |
| OpenAI | 1.92 |
| Claude Sonnet 5.5 | 1.5 |
| AMD | 1.3 |
| Nvidia | 1.26 |
| Opus 5.5 | 1.25 |
Heat = summed hotness of today's stories naming the topic.
Models
5 storiesReleases, papers, trending weights and new listings
Top 6 trending on Hugging Face: likes
Table
| convaiinnovations/laya | 4298 |
| Qwen/Qwen-Image-2.1 | 2588 |
| abenzerps/Qwen-Image-2.1-Uncensored-GGUF | 2259 |
| XingChen-AGI/Xing4.0-29B-A4B | 1799 |
| Edge0/Audio8-ASR-Infinite | 1375 |
| XingChen-AGI/TeleOCR | 770 |
PirateBench: score out of 100
Table
| opus-5-5 | 91.15 |
| gpt-6-sol | 81.78 |
| deepseek-v4-flash-0731 | 62.82 |
Agents build our own projects from a brief; every run, failures included: https://cowerx.dev/daily/piratebench/
Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms
141 points and 39 comments on Hacker News, linking github.com.
Claude Sonnet 5.5
191 points and 138 comments on Hacker News, linking anthropic.com.
Alissonerdx/BFS-Best-Face-Swap
A image to image model from Alissonerdx with 1,010 likes and 150.0k downloads on Hugging Face.
Launch HN: Vespper (YC F24) – SOTA Docx MCP
Vespper released a model fine-tuned for editing Word documents as an MCP. It reportedly makes agents three times faster and twice as cheap as the closest alternative. ·
Claude Sonnet 5.5 in GitHub Copilot
Anthropic made its Claude Sonnet 5.5 model generally available in GitHub Copilot. The model targets routine development tasks such as building features and fixing bugs. ·
Claude Sonnet 5.5
191 points and 138 comments on Hacker News, linking anthropic.com.
Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms
141 points and 39 comments on Hacker News, linking github.com.
unslothai/unsloth Laya Decision Models + Library
Unsloth added decision model support, a document library, and Apple Silicon optimizations. The update also enables local Laya execution and accelerates image and video generation by up to 4.5 times. ·
Launch HN: Vespper (YC F24) – SOTA Docx MCP
Vespper released a model fine-tuned for editing Word documents as an MCP. It reportedly makes agents three times faster and twice as cheap as the closest alternative. ·
Alissonerdx/BFS-Best-Face-Swap
A image to image model from Alissonerdx with 1,010 likes and 150.0k downloads on Hugging Face.
Claude Sonnet 5.5
191 points and 138 comments on Hacker News, linking anthropic.com.
Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms
141 points and 39 comments on Hacker News, linking github.com.
unslothai/unsloth Laya Decision Models + Library
Unsloth added decision model support, a document library, and Apple Silicon optimizations. The update also enables local Laya execution and accelerates image and video generation by up to 4.5 times. ·
AI companies in race to demonstrate their model most threatening to humanity
424 points and 382 comments on Hacker News, linking thecivilian.co.nz.
Disaggregated Quantization: Specializing LLM Prefill and Decode
Researchers introduced disaggregated quantization, which uses separate low-precision formats for LLM prompt processing and token generation. This approach improved 1-bit accuracy by 32.5 points on MMLU-Pro and sped up initial token generation by 1.78 times. ·
X voices
5 storiesX posts other sources are linking to, then named voices on Bluesky and Mastodon
@thesupermannx
Researchers proved AI has deleted every reason universities exist
1 points and 3 comments on Hacker News, linking an X post.
@drfeifei
AMD Acquires Dr. Fei Fei Li's World Labs
11 points and 1 comments on Hacker News, linking an X post.
@danielhayesmith
I built a native Apple app for 4 platforms in one afternoon with my AI workflow
9 points and 2 comments on Hacker News, linking an X post.
@HedgieMarkets
Hospitals use AI to find more things to bill for. Insurers use AI to deny them.
12 points and 1 comments on Hacker News, linking an X post.
@drfeifei
AMD Acquires Dr. Fei Fei Li's World Labs
11 points and 1 comments on Hacker News, linking an X post.
@thesupermannx
Researchers proved AI has deleted every reason universities exist
1 points and 3 comments on Hacker News, linking an X post.
@EdLudlow
AMD is acquiring Fei-Fei Li's World Labs in an $8.2B, all-stock deal
2 points and 1 comments on Hacker News, linking an X post.
@danielhayesmith
I built a native Apple app for 4 platforms in one afternoon with my AI workflow
9 points and 2 comments on Hacker News, linking an X post.
@drfeifei
AMD Acquires Dr. Fei Fei Li's World Labs
11 points and 1 comments on Hacker News, linking an X post.
@thesupermannx
Researchers proved AI has deleted every reason universities exist
1 points and 3 comments on Hacker News, linking an X post.
@danielhayesmith
I built a native Apple app for 4 platforms in one afternoon with my AI workflow
9 points and 2 comments on Hacker News, linking an X post.
@EdLudlow
AMD is acquiring Fei-Fei Li's World Labs in an $8.2B, all-stock deal
2 points and 1 comments on Hacker News, linking an X post.
GitHub
5 storiesNew repos, releases and changelog
New AI repos this week: stars
Table
| yetone/magpie | 1584 |
| dzhng/jevgrep | 1389 |
| mikehasa/golive-skill | 1044 |
| ollaya-dev/ollaya | 874 |
| feitangyuan/onetake | 820 |
| yukitorido/short-video-generator-AI | 731 |
Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms
141 points and 39 comments on Hacker News, linking github.com.
langchain==1.4.3
LangChain released version 1.4.3, adding support for Bedrock Mantle chat models and fixing cache settings for fallback models. ·
Claude Sonnet 5.5 in GitHub Copilot
Anthropic made its Claude Sonnet 5.5 model generally available in GitHub Copilot. The model targets routine development tasks such as building features and fixing bugs. ·
anthropics/anthropic-sdk-python v1.9.0
New release v1.9.0 of anthropics/anthropic-sdk-python on GitHub.
unslothai/unsloth Laya Decision Models + Library
Unsloth added decision model support, a document library, and Apple Silicon optimizations. The update also enables local Laya execution and accelerates image and video generation by up to 4.5 times. ·
Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms
141 points and 39 comments on Hacker News, linking github.com.
unslothai/unsloth Laya Decision Models + Library
Unsloth added decision model support, a document library, and Apple Silicon optimizations. The update also enables local Laya execution and accelerates image and video generation by up to 4.5 times. ·
openai/codex 0.158.0
OpenAI released version 0.158.0 of Codex, adding features for terminal interaction, OAuth connections, and image handling. The update also includes fixes for sandbox issues on Windows, Linux, and macOS. ·
JoasASantos/Offensive-Security-AI-Models
JoasASantos maintains a repository of AI models that are uncensored or fine-tuned for cybersecurity tasks. ·
langchain==1.4.3
LangChain released version 1.4.3, adding support for Bedrock Mantle chat models and fixing cache settings for fallback models. ·
Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms
141 points and 39 comments on Hacker News, linking github.com.
unslothai/unsloth Laya Decision Models + Library
Unsloth added decision model support, a document library, and Apple Silicon optimizations. The update also enables local Laya execution and accelerates image and video generation by up to 4.5 times. ·
openai/codex 0.158.0
OpenAI released version 0.158.0 of Codex, adding features for terminal interaction, OAuth connections, and image handling. The update also includes fixes for sandbox issues on Windows, Linux, and macOS. ·
JoasASantos/Offensive-Security-AI-Models
JoasASantos maintains a repository of AI models that are uncensored or fine-tuned for cybersecurity tasks. ·
zhuyansen/awesome-claude-video-skills
A developer published a curated list of 180 open-source repositories. These tools allow coding agents to generate video and include security grades. ·
Data centers & energy
5 storiesBuilds, power deals and the grid
Data-center and energy stories per day
Table
| Tue 22 | 3 |
| Wed 23 | 3 |
| Thu 24 | 3 |
| Fri 25 | 20 |
| Sat 26 | 1 |
| Sun 27 | 1 |
| Mon 28 | 20 |
Stories from DCD, Data Center Knowledge, EIA and Hacker News, by publication day.
Goodman withdraws application for 90MW data center campus in Sydney, Australia
Goodman withdrew its application for a 90-megawatt data center campus in Sydney, citing significant changes to the local policy and regulatory environment. ·
Re-engineering silicon and system design for AI-era workloads
The piece examines the design of silicon and systems required to support current and future data center workloads for artificial intelligence. ·
Can Supercapacitors Smooth AI’s Power Swings?
Supercapacitors offer millisecond response to AI power fluctuations. Experts view them as complements to existing battery and UPS systems rather than replacements. ·
Every Household in This Rural Town Receives $10k If a Data Center Gets Built
10 points and 5 comments on Hacker News, linking wsj.com.
Goodman withdraws application for 90MW data center campus in Sydney, Australia
Goodman withdrew its application for a 90-megawatt data center campus in Sydney, citing significant changes to the local policy and regulatory environment. ·
Re-engineering silicon and system design for AI-era workloads
The piece examines the design of silicon and systems required to support current and future data center workloads for artificial intelligence. ·
Can Supercapacitors Smooth AI’s Power Swings?
Supercapacitors offer millisecond response to AI power fluctuations. Experts view them as complements to existing battery and UPS systems rather than replacements. ·
Every Household in This Rural Town Receives $10k If a Data Center Gets Built
10 points and 5 comments on Hacker News, linking wsj.com.
Goodman withdraws application for 90MW data center campus in Sydney, Australia
Goodman withdrew its application for a 90-megawatt data center campus in Sydney, citing significant changes to the local policy and regulatory environment. ·
Re-engineering silicon and system design for AI-era workloads
The piece examines the design of silicon and systems required to support current and future data center workloads for artificial intelligence. ·
Can Supercapacitors Smooth AI’s Power Swings?
Supercapacitors offer millisecond response to AI power fluctuations. Experts view them as complements to existing battery and UPS systems rather than replacements. ·
Hardware prices
2 storiesGPUs, memory, and what tokens cost
Output price, $ per million tokens (newest models)
Table
| perceptron-mk1.5 | 1.5 |
| ember-1 | 15 |
| glm-5.3-prime | 8.8 |
| qwen3.8-max-prime | 12 |
| aion-3.5-mini | 1.4 |
| median, all paid models | 2 |
432 paid models listed; median of the 54 added in 30 days: $2.85. Rankings data by OpenRouter, CC BY 4.0.
Import AI 474: Platonic mindspace; TPUs in space; Zhipu starts an outer RSI loop
Import AI issue 474 covers Platonic mindspace, space-based TPUs, and Zhipu's outer RSI loop. ·
From Import AI.
Import AI 474: Platonic mindspace; TPUs in space; Zhipu starts an outer RSI loop
Import AI issue 474 covers Platonic mindspace, space-based TPUs, and Zhipu's outer RSI loop. ·
From Import AI.
Import AI 474: Platonic mindspace; TPUs in space; Zhipu starts an outer RSI loop
Import AI issue 474 covers Platonic mindspace, space-based TPUs, and Zhipu's outer RSI loop. ·
From Import AI.
Virtio-nvgpu: Near-native Nvidia GPU access inside a KVM guest
An experimental project provides a virtio device that allows KVM virtual machines to access NVIDIA GPUs with near-native performance. ·
