Archived front page, Tue 29 Sep 2026, 15:00 UTC. All of 2026-09-29 · Archive · Live page

The Daily Markdown

The front page of the internet for AI users · Tue 29 Sep 2026, 15:00 UTC

Top now

6 stories

What is climbing fastest across every source

OpenAI6.5xWorld Labs7.5xAMD7.3xLearning6.7xAnthropic6.9xClaude Sonnet 5.57.0xGeneral Agents6.8xLanguage6.2xOpus 5.57.0xDevDay7.1x

Claude partial outage

67 points and 32 comments on Hacker News, linking status.claude.com.

Hacker News34m ago▲ 7545 to 99 across 2 pollsdiscussstatus.claude.com

PostHog/jeeves

PostHog released a project named Jeeves that uses reasoning to improve Jev-like decision models. ·

GitHub+ Hacker News5h ago▲ 17github.com

PostHog/jeeves

PostHog released a project named Jeeves that uses reasoning to improve Jev-like decision models. ·

A new repository written in Python with 116 stars in its first week.

GitHub+ Hacker News5h ago▲ 174github.com

Claude partial outage

67 points and 32 comments on Hacker News, linking status.claude.com.

Hacker News34m ago▲ 9745 to 99 across 2 pollsdiscussstatus.claude.com

openai/codex 0.159.0

OpenAI released version 0.159.0 of Codex, adding an opt-in feature that allows new input to steer the model during active responses or long-running code calls. ·

GitHub Releases7h ago▲ 7513 to 23 across 3 pollsgithub.com

Groupwise Agentic Grading and Advantage Redistribution for Code Agent RL

Researchers introduced GAGAR, a framework that uses an agentic grader to rank code agent trajectories. It redistributes reinforcement learning advantages to favor higher-quality implementations over those with unnecessary changes. ·

HF Daily Papers15h ago▲ 537 to 58 across 5 pollsarxiv.org

PostHog/jeeves

PostHog released a project named Jeeves that uses reasoning to improve Jev-like decision models. ·

A new repository written in Python with 116 stars in its first week.

GitHub+ Hacker News5h ago▲ 201github.com

Claude Sonnet 5.5

191 points and 138 comments on Hacker News, linking anthropic.com.

Hacker News+ Simon Willison21h ago▲ 11665 to 329 across 4 pollsdiscussanthropic.com

JoasASantos/Offensive-Security-AI-Models

JoasASantos maintains a repository of AI models that are uncensored or fine-tuned for cybersecurity tasks. ·

GitHub+ Hacker News42h ago▲ 112110 to 341 across 24 pollsgithub.com

Claude partial outage

67 points and 32 comments on Hacker News, linking status.claude.com.

Hacker News34m ago▲ 9945 to 99 across 2 pollsdiscussstatus.claude.com

openai/codex 0.159.0

OpenAI released version 0.159.0 of Codex, adding an opt-in feature that allows new input to steer the model during active responses or long-running code calls. ·

GitHub Releases7h ago▲ 9113 to 23 across 3 pollsgithub.com

Trending topics today: heat

OpenAI: 3.33OpenAI3.33World Labs: 1.75World Labs1.75AMD: 1.28AMD1.28Learning: 1.10Learning1.10Anthropic: 0.90Anthropic0.90Claude Sonnet 5.5: 0.90Claude Sonnet 5.50.90
Table
OpenAI3.33
World Labs1.75
AMD1.28
Learning1.1
Anthropic0.9
Claude Sonnet 5.50.9

Heat = summed hotness of today's stories naming the topic.

Models

5 stories

Releases, papers, trending weights and new listings

Top 6 trending on Hugging Face: likes

convaiinnovations/laya: 4,450convaiinnovations/laya4,450Qwen/Qwen-Image-2.1: 2,638Qwen/Qwen-Image-2.12,638abenzerps/Qwen-Image-2.1-Uncensored-GGUF: 2,374abenzerps/Qwen-Image-…2,374XingChen-AGI/Xing4.0-29B-A4B: 1,806XingChen-AGI/Xing4.0-…1,806Edge0/Audio8-ASR-Infinite: 1,470Edge0/Audio8-ASR-Infi…1,470XingChen-AGI/TeleOCR: 855XingChen-AGI/TeleOCR855
Table
convaiinnovations/laya4450
Qwen/Qwen-Image-2.12638
abenzerps/Qwen-Image-2.1-Uncensored-GGUF2374
XingChen-AGI/Xing4.0-29B-A4B1806
Edge0/Audio8-ASR-Infinite1470
XingChen-AGI/TeleOCR855

PirateBench: score out of 100

opus-5-5: 91opus-5-591gpt-6-sol: 82gpt-6-sol82deepseek-v4-flash-0731: 63deepseek-v4-flash-073163
Table
opus-5-591.15
gpt-6-sol81.78
deepseek-v4-flash-073162.82

Agents build our own projects from a brief; every run, failures included: https://cowerx.dev/daily/piratebench/

Language Models for Text Classification: From Bag-of-Words to Jev

The source presents a visual guide to recurrent neural networks, convolutional neural networks, and transformers. It includes hands-on experiments evaluating model accuracy and efficiency for text classification tasks. ·

Ahead of AI4h ago▲ 4magazine.sebastianraschka.com

Claude Sonnet 5.5

191 points and 138 comments on Hacker News, linking anthropic.com.

Hacker News+ Simon Willison21h ago▲ 6565 to 329 across 4 pollsdiscussanthropic.com

Groupwise Agentic Grading and Advantage Redistribution for Code Agent RL

Researchers introduced GAGAR, a framework that uses an agentic grader to rank code agent trajectories. It redistributes reinforcement learning advantages to favor higher-quality implementations over those with unnecessary changes. ·

HF Daily Papers15h ago▲ 537 to 58 across 5 pollsarxiv.org

Claude Sonnet 5.5

191 points and 138 comments on Hacker News, linking anthropic.com.

Hacker News+ Simon Willison21h ago▲ 11665 to 329 across 4 pollsdiscussanthropic.com

JoasASantos/Offensive-Security-AI-Models

JoasASantos maintains a repository of AI models that are uncensored or fine-tuned for cybersecurity tasks. ·

GitHub+ Hacker News42h ago▲ 112110 to 341 across 24 pollsgithub.com

Groupwise Agentic Grading and Advantage Redistribution for Code Agent RL

Researchers introduced GAGAR, a framework that uses an agentic grader to rank code agent trajectories. It redistributes reinforcement learning advantages to favor higher-quality implementations over those with unnecessary changes. ·

HF Daily Papers15h ago▲ 817 to 58 across 5 pollsarxiv.org

X voices

5 stories

X posts other sources are linking to, then named voices on Bluesky and Mastodon

GitHub

5 stories

New repos, releases and changelog

New AI repos this week: stars

KKKKhazix/AIHOT: 2,624KKKKhazix/AIHOT2,624yetone/magpie: 2,137yetone/magpie2,137dzhng/jevgrep: 1,670dzhng/jevgrep1,670mikehasa/golive-skill: 1,092mikehasa/golive-skill1,092ollaya-dev/ollaya: 946ollaya-dev/ollaya946feitangyuan/onetake: 945feitangyuan/onetake945
Table
KKKKhazix/AIHOT2624
yetone/magpie2137
dzhng/jevgrep1670
mikehasa/golive-skill1092
ollaya-dev/ollaya946
feitangyuan/onetake945

openai/codex 0.159.0

OpenAI released version 0.159.0 of Codex, adding an opt-in feature that allows new input to steer the model during active responses or long-running code calls. ·

New release rust-v0.159.0 of openai/codex on GitHub.

GitHub Releases7h ago▲ 313 to 23 across 3 pollsgithub.com

KKKKhazix/AIHOT

A website framework automatically identifies trending topics and generates daily reports. Users can customize its sources and selection criteria to create industry-specific news sites. ·

GitHub15h ago▲ 0462 to 2624 across 14 pollsgithub.com

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. ·

GitHub Releases18h ago▲ 0github.com

openai/codex 0.159.0

OpenAI released version 0.159.0 of Codex, adding an opt-in feature that allows new input to steer the model during active responses or long-running code calls. ·

New release rust-v0.159.0 of openai/codex on GitHub.

GitHub Releases7h ago▲ 7513 to 23 across 3 pollsgithub.com

KKKKhazix/AIHOT

A website framework automatically identifies trending topics and generates daily reports. Users can customize its sources and selection criteria to create industry-specific news sites. ·

GitHub15h ago▲ 53462 to 2624 across 14 pollsgithub.com

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. ·

GitHub Releases21h ago▲ 4030 to 50 across 10 pollsgithub.com

JoasASantos/Offensive-Security-AI-Models

JoasASantos maintains a repository of AI models that are uncensored or fine-tuned for cybersecurity tasks. ·

A new repository with 341 stars in its first week.

GitHub+ Hacker News42h ago▲ 112110 to 341 across 24 pollsgithub.com

openai/codex 0.159.0

OpenAI released version 0.159.0 of Codex, adding an opt-in feature that allows new input to steer the model during active responses or long-running code calls. ·

GitHub Releases7h ago▲ 9113 to 23 across 3 pollsgithub.com

KKKKhazix/AIHOT

A website framework automatically identifies trending topics and generates daily reports. Users can customize its sources and selection criteria to create industry-specific news sites. ·

GitHub15h ago▲ 81462 to 2624 across 14 pollsgithub.com

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. ·

GitHub Releases21h ago▲ 7230 to 50 across 10 pollsgithub.com

Data centers & energy

5 stories

Builds, power deals and the grid

Data-center and energy stories per day

Wed 23: 3Wed 233Thu 24: 3Thu 243Fri 25: 20Fri 2520Sat 26: 1Sat 261Sun 27: 1Sun 271Mon 28: 20Mon 2820Tue 29: 18Tue 2918
Table
Wed 233
Thu 243
Fri 2520
Sat 261
Sun 271
Mon 2820
Tue 2918

Stories from DCD, Data Center Knowledge, EIA and Hacker News, by publication day.

Hardware prices

2 stories

GPUs, memory, and what tokens cost

Output price, $ per million tokens (newest models)

claude-sonnet-5.5: 10claude-sonnet-5.510claude-sonnet-5.5:batch: 5claude-sonnet-5.5:bat…5perceptron-mk1.5: 1.50perceptron-mk1.51.50ember-1: 15ember-115glm-5.3-prime: 8.80glm-5.3-prime8.80median, all paid models: 2median, all paid mode…2
Table
claude-sonnet-5.510
claude-sonnet-5.5:batch5
perceptron-mk1.51.5
ember-115
glm-5.3-prime8.8
median, all paid models2

434 paid models listed; median of the 56 added in 30 days: $4.00. Rankings data by OpenRouter, CC BY 4.0.