Archived front page, Tue 29 Sep 2026, 18:00 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, 18:00 UTC

Top now

6 stories

What is climbing fastest across every source

OpenAI6.7xClaude7.0xAnthropic7.0xDevDay7.0xWorld Labs7.5xAMD7.3xMeta6.9xCodex7.7xLanguage6.4xLearning6.6x

ChatGPT Pro 500

61 points and 45 comments on Hacker News, linking help.openai.com.

Hacker News34m ago▲ 745 to 106 across 3 pollsdiscusshelp.openai.com

GPT 6.1 Sol

268 points and 187 comments on Hacker News, linking openai.com.

Hacker News53m ago▲ 6471 to 455 across 5 pollsdiscussopenai.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 198 stars in its first week.

GitHub+ Hacker News8h ago▲ 153116 to 198 across 4 pollsgithub.com

GPT 6.1 Sol

268 points and 187 comments on Hacker News, linking openai.com.

Hacker News53m ago▲ 9671 to 455 across 5 pollsdiscussopenai.com

ChatGPT Pro 500

61 points and 45 comments on Hacker News, linking help.openai.com.

Hacker News34m ago▲ 965 to 106 across 3 pollsdiscusshelp.openai.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 Releases10h ago▲ 6613 to 27 across 5 pollsgithub.com

Post-Training Leaves Behavioral Shadows on Unrelated Decisions

Researchers introduced Active Taskless Distillation, a method transferring capabilities using single-word prompts. In coding tests, this approach yielded a 5.34 percentage point gain on HumanEval+. ·

HF Daily Papers18h ago▲ 4711 to 193 across 12 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 198 stars in its first week.

GitHub+ Hacker News8h ago▲ 192116 to 198 across 4 pollsgithub.com

Claude Sonnet 5.5

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

Hacker News+ Simon Willison24h ago▲ 11165 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 News45h ago▲ 108110 to 363 across 27 pollsgithub.com

GPT 6.1 Sol

268 points and 187 comments on Hacker News, linking openai.com.

Hacker News53m ago▲ 9971 to 455 across 5 pollsdiscussopenai.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 Releases10h ago▲ 8713 to 27 across 5 pollsgithub.com

Post-Training Leaves Behavioral Shadows on Unrelated Decisions

Researchers introduced Active Taskless Distillation, a method transferring capabilities using single-word prompts. In coding tests, this approach yielded a 5.34 percentage point gain on HumanEval+. ·

HF Daily Papers18h ago▲ 7811 to 193 across 12 pollsarxiv.org

Trending topics today: heat

OpenAI: 4.69OpenAI4.69Claude: 1.75Claude1.75Anthropic: 1.66Anthropic1.66DevDay: 1.65DevDay1.65World Labs: 1.52World Labs1.52AMD: 1.10AMD1.10
Table
OpenAI4.69
Claude1.75
Anthropic1.66
DevDay1.65
World Labs1.52
AMD1.1

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: 4,477convaiinnovations/laya4,477Qwen/Qwen-Image-2.1: 2,644Qwen/Qwen-Image-2.12,644abenzerps/Qwen-Image-2.1-Uncensored-GGUF: 2,392abenzerps/Qwen-Image-…2,392Edge0/Audio8-ASR-Infinite: 1,479Edge0/Audio8-ASR-Infi…1,479XingChen-AGI/TeleOCR: 860XingChen-AGI/TeleOCR860Contrastive-LM/CLM-v0.1-8B: 516Contrastive-LM/CLM-v0…516
Table
convaiinnovations/laya4477
Qwen/Qwen-Image-2.12644
abenzerps/Qwen-Image-2.1-Uncensored-GGUF2392
Edge0/Audio8-ASR-Infinite1479
XingChen-AGI/TeleOCR860
Contrastive-LM/CLM-v0.1-8B516

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/

Claude partial outage

165 points and 136 comments on Hacker News, linking status.claude.com.

Hacker News4h ago▲ 1545 to 301 across 9 pollsdiscussstatus.claude.com

Claude partial outage

165 points and 136 comments on Hacker News, linking status.claude.com.

Hacker News4h ago▲ 8345 to 301 across 9 pollsdiscussstatus.claude.com

Claude Sonnet 5.5

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

Hacker News+ Simon Willison24h ago▲ 5765 to 329 across 4 pollsdiscussanthropic.com

Post-Training Leaves Behavioral Shadows on Unrelated Decisions

Researchers introduced Active Taskless Distillation, a method transferring capabilities using single-word prompts. In coding tests, this approach yielded a 5.34 percentage point gain on HumanEval+. ·

HF Daily Papers18h ago▲ 4711 to 193 across 12 pollsarxiv.org

Claude Sonnet 5.5

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

Hacker News+ Simon Willison24h ago▲ 11165 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 News45h ago▲ 108110 to 363 across 27 pollsgithub.com

Claude partial outage

165 points and 136 comments on Hacker News, linking status.claude.com.

Hacker News4h ago▲ 9145 to 301 across 9 pollsdiscussstatus.claude.com

Post-Training Leaves Behavioral Shadows on Unrelated Decisions

Researchers introduced Active Taskless Distillation, a method transferring capabilities using single-word prompts. In coding tests, this approach yielded a 5.34 percentage point gain on HumanEval+. ·

HF Daily Papers18h ago▲ 7811 to 193 across 12 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: 2,869KKKKhazix/AIHOT2,869yetone/magpie: 2,482yetone/magpie2,482dzhng/jevgrep: 1,716dzhng/jevgrep1,716mikehasa/golive-skill: 1,100mikehasa/golive-skill1,100feitangyuan/onetake: 962feitangyuan/onetake962ollaya-dev/ollaya: 958ollaya-dev/ollaya958
Table
KKKKhazix/AIHOT2869
yetone/magpie2482
dzhng/jevgrep1716
mikehasa/golive-skill1100
feitangyuan/onetake962
ollaya-dev/ollaya958

langchain-anthropic==1.7.5

The langchain-anthropic library released version 1.7.5, adding support for Claude Sonnet 5.5 and fixing how invalid tool calls are handled during replay. ·

New release langchain-anthropic==1.7.5 of langchain-ai/langchain on GitHub.

GitHub Releases3h ago▲ 8github.com

langchain-core==1.6.6

LangChain released version 1.6.6, adding support for Claude Sonnet 5.5 compatibility and correcting docstring examples that failed to run when copied. ·

GitHub Releases3h ago▲ 6github.com

PostHog/jeeves

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

GitHub+ Hacker News8h ago▲ 4116 to 198 across 4 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 Releases10h ago▲ 113 to 27 across 5 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. ·

GitHub18h ago▲ 0462 to 2869 across 17 pollsgithub.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 198 stars in its first week.

GitHub+ Hacker News8h ago▲ 153116 to 198 across 4 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 Releases10h ago▲ 6613 to 27 across 5 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. ·

GitHub18h ago▲ 47462 to 2869 across 17 pollsgithub.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 198 stars in its first week.

GitHub+ Hacker News8h ago▲ 192116 to 198 across 4 pollsgithub.com

JoasASantos/Offensive-Security-AI-Models

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

GitHub+ Hacker News45h ago▲ 108110 to 363 across 27 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 Releases10h ago▲ 8713 to 27 across 5 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. ·

GitHub18h ago▲ 78462 to 2869 across 17 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 Releases24h ago▲ 7030 to 50 across 12 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: 32Tue 2932
Table
Wed 233
Thu 243
Fri 2520
Sat 261
Sun 271
Mon 2820
Tue 2932

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

Free space optics: To the moon and back

Researchers are exploring optical communication for space applications. The technology's success depends on its ability to function reliably despite adverse weather conditions. ·

DCD2h ago▲ 11datacenterdynamics.com

Free space optics: To the moon and back

Researchers are exploring optical communication for space applications. The technology's success depends on its ability to function reliably despite adverse weather conditions. ·

DCD2h ago▲ 28datacenterdynamics.com

Free space optics: To the moon and back

Researchers are exploring optical communication for space applications. The technology's success depends on its ability to function reliably despite adverse weather conditions. ·

DCD2h ago▲ 29datacenterdynamics.com

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.