The front page of the internet for AI users · Fri 06 Nov 2026, 00:00 UTC · Get it by email
What is climbing fastest across every source
Nasdaq and chip stocks fell after voters rejected data centers, prompting three scenarios for AI builders. ·
3 points and 1 comments on Hacker News, linking vithushansylvester.com.
Claude Code sends user email addresses to websites in the User-Agent request header during web searches without permission. ·
webAI-Official releases TwIL-8B, a text generation transformer model with 9B parameters available on Hugging Face. ·
Langchain-typesafe version 0.0.2 accepts a classifier in experimental middleware. ·
Sentinel Bio co-founder Joshua Monrad discusses DNA screening, uplift studies, and agent monitoring with Nathan Labenz. ·
From Dwarkesh Podcast.
Claude Code sends user email addresses to websites in the User-Agent request header during web searches without permission. ·
255 points and 71 comments on Hacker News, linking github.com.
69 points and 1 comments on Hacker News, linking tencent.com.
webAI-Official releases TwIL-8B, a text generation transformer model with 9B parameters available on Hugging Face. ·
Gemini Omni 1.2 Flash offers native 15-second video generation and editing with stronger consistency and 30% lower pricing. ·
Kimi K3.1 Pro is an image-text-to-text model with 2.8 trillion parameters available on Hugging Face. ·
Multi-teacher on-policy distillation consolidates capabilities from multiple teachers but exhibits incomplete generalization beyond the supervised distribution. ·
Kimi K3.1 Pro is an image-text-to-text model with 2.8 trillion parameters available on Hugging Face. ·
617 points and 184 comments on Hacker News, linking huggingface.co.
OpenAI acquires Pierre Computer Company to expand Codex tools for building, understanding, and reviewing code. ·
Gemini Omni 1.2 Flash offers native 15-second video generation and editing with stronger consistency and 30% lower pricing. ·
TurnWise applies turn-level Group Relative Policy Optimization to streaming video understanding, separating credit assignment and averaging token losses to improve decision making. ·
Multi-teacher on-policy distillation consolidates capabilities from multiple teachers but exhibits incomplete generalization beyond the supervised distribution. ·
ggml-org releases llama.cpp v0.9.0, adding OCP FP8 E4M3 quantization, flash attention support, and four new model architectures. ·
| OpenAI | 1.8 |
| GPT-6 Cyber | 1.38 |
| Gina Hinojosa | 1.1 |
| Claude Code | 0.93 |
| GPUs | 0.9 |
| Gemini Omni 1.2 Flash | 0.86 |
Heat = summed hotness of today's stories naming the topic.
Releases, papers, trending weights and new listings
| Lightricks/LTX-2.5 ▲5 | 10724 |
| mistralai/Mistral-Large-4-1T-A52B ▼1 | 1613 |
| moonshotai/Kimi-K3.1-Pro ▲5 | 915 |
| speridlabs/idol ▲1 | 766 |
| ccore/Qwen3.5-4B-3D-Coder ▲4 | 700 |
| Pika-Labs/Pika-H3-Flex ▲3 | 473 |
| gpt-6.1-sol | 98.9 |
| opus-5-5 | 96.46 |
| gpt-6-sol | 91.6 |
| gpt-6-luna | 65.25 |
| haiku-4-5 | 4.28 |
Agents build our own projects from a brief; every run, failures included: https://cowerx.dev/daily/piratebench/
69 points and 1 comments on Hacker News, linking tencent.com.
Hungary released communist-era secret police files on Wednesday, identifying up to 60,000 informants. ·
70 points and 37 comments on Hacker News, linking github.com.
Gemini Omni 1.2 Flash offers native 15-second video generation and editing with stronger consistency and 30% lower pricing. ·
Kimi K3.1 Pro is an image-text-to-text model with 2.8 trillion parameters available on Hugging Face. ·
Kimi K3.1 Pro is an image-text-to-text model with 2.8 trillion parameters available on Hugging Face. ·
617 points and 184 comments on Hacker News, linking huggingface.co.
Gemini Omni 1.2 Flash offers native 15-second video generation and editing with stronger consistency and 30% lower pricing. ·
TurnWise applies turn-level Group Relative Policy Optimization to streaming video understanding, separating credit assignment and averaging token losses to improve decision making. ·
A paper by Rikiya Takehi, Jayden Teoh, Alex Zhang, and Omar Khattab describes weight-space programming, where models like GPT-6 Astra read and write neural network weights. ·
Hungary released communist-era secret police files on Wednesday, identifying up to 60,000 informants. ·
X posts other sources are linking to, then named voices on Bluesky and Mastodon
Frameboy: Explore worlds in real-time streaming from cloud GPUs at up to 15 FPS
2 points and 0 comments on Hacker News, linking an X post.
Z.ai says GLM 5.5 built on 1T web pages
7 points and 0 comments on Hacker News, linking an X post.
Taylor Lorenz on X: "I'm once again being offered money promote anti-AI content
4 points and 0 comments on Hacker News, linking an X post.
OpenAI spent $1.2M in API cost to fix 948 Linux vulnerabilities
4 points and 0 comments on Hacker News, linking an X post.
Pierre Computer is being acquired by OpenAI
72 points and 32 comments on Hacker News, linking an X post.
Z.ai says GLM 5.5 built on 1T web pages
7 points and 0 comments on Hacker News, linking an X post.
Yassamin Ansari: Top of the Democratic agenda on day 1 pause on advanced AI dev
4 points and 10 comments on Hacker News, linking an X post.
Researcher accuses Anthropic of stealing his work
10 points and 1 comments on Hacker News, linking an X post.
New repos, releases and changelog
| deepseek-ai/dsh-superagent-prototype | 19458 |
| KKKKhazix/agent-autoloop | 3220 |
| worldonlyu/Google_AI_Pro_2Year_ZeroCost_Auto | 2282 |
| caru-ini/stack-diagram | 1863 |
| wquguru/cicada | 1478 |
| N1nEmAn/sapper-v2 | 1327 |
70 points and 37 comments on Hacker News, linking github.com.
A USB-based offline survival assistant uses a local LLM and Kiwix to provide guides on Windows, Linux, and macOS. ·
Stroq is an open source tool that audits local coding agent sessions to find secrets and guard future ones with local hooks. ·
Majorminorlabs releases Hermes Voice, an agent profile that allows users to talk through Home Assistant Assist. ·
Unsloth releases Qwen3.8-Flash-Next, updates Glass themes, adds a Composer position slider, and enables browser file interaction with lower idle memory usage. ·
70 points and 37 comments on Hacker News, linking github.com.
ggml-org releases llama.cpp v0.9.0, adding OCP FP8 E4M3 quantization, flash attention support, and four new model architectures. ·
A new repository written in Python with 2,282 stars in its first week.
Stroq is an open source tool that audits local coding agent sessions to find secrets and guard future ones with local hooks. ·
Unsloth releases Qwen3.8-Flash-Next, updates Glass themes, adds a Composer position slider, and enables browser file interaction with lower idle memory usage. ·
Data centers, power, chips and what tokens cost
| Sat 31 | 2 |
| Sun 01 | 5 |
| Mon 02 | 34 |
| Tue 03 | 39 |
| Wed 04 | 40 |
| Thu 05 | 36 |
| Fri 06 | 0 |
Stories from DCD, Data Center Knowledge, EIA and Hacker News, by publication day.
One point per OpenRouter reading, last 8 days. Data by OpenRouter, CC BY 4.0.
| gemini-omni-1.2-flash | 9 |
| gemini-omni-1.1-flash | 9 |
| ripple-65b-preview | 0.6 |
| hy4 | 1.2505 |
| glm-5.3-flashx-legal | 1.25 |
| median, all paid models | 2.09 |
428 paid models listed; median of the 31 added in 30 days: $2.50. Rankings data by OpenRouter, CC BY 4.0.
Halley is a predictive control plane for optical AI data center networks that improves effective goodput from 65.2% to 76.5%. ·
27 upvotes on Hugging Face Daily Papers.
The SIMPLE law predicts LLM inference energy using four parameters, achieving high accuracy on NVIDIA H100 GPUs without per-model calibration. ·
Qualcomm Cloud AI 100 Ultra is a 150W AI accelerator card with 128GB memory and 548GB/s bandwidth that accelerates generative AI. ·
US voters in over 30 localities across nine states reject data centers through ballot measures including moratoriums and bans. ·
344 points and 737 comments on Hacker News, linking semafor.com.
Halley is a predictive control plane for optical AI data center networks that improves effective goodput from 65.2% to 76.5%. ·
Qualcomm Cloud AI 100 Ultra is a 150W AI accelerator card with 128GB memory and 548GB/s bandwidth that accelerates generative AI. ·
The SIMPLE law predicts LLM inference energy using four parameters, achieving high accuracy on NVIDIA H100 GPUs without per-model calibration. ·
Tech stocks decline in overnight trading as the probability of a blue wave increases following the Texas governor result. ·
