Archived front page, Wed 30 Sep 2026, 22:00 UTC. All of 2026-09-30 · Archive · Live page

The Daily Markdown

The front page of the internet for AI users · Wed 30 Sep 2026, 22:00 UTC

Top now

6 stories

What is climbing fastest across every source

Gemini 4 Argon7.5xOpenAI5.8xAnthropic6.1xDots7.1xLLM6.6xGoogle7.0xGPT-6.17.1xAstra7.0xGPT-6.1 Sol7.1xPro8.5x

Data Center On-Site Power Brings Pollution Risks Closer to Home

Data center operators increasingly use on-site power generation to bypass grid constraints, creating new environmental compliance challenges and bringing pollution risks closer to local communities. ·

Data Center Knowledge4h ago▲datacenterknowledge.com

Gemini 4 Argon

644 points and 404 comments on Hacker News, linking blog.google.

Hacker News2h ago▲51 to 1048 across 10 pollsdiscussblog.google

PSRben/VisionHOPE

A image classification model from PSRben with 226 likes and 137 downloads on Hugging Face.

Hugging Face5h ago▲162 to 226 across 11 pollshuggingface.co

pyannote/speaker-diarization-community-1

A automatic speech recognition model from pyannote with 2,340 likes and 5.5M downloads on Hugging Face.

Hugging Face8h ago▲2316 to 2340 across 17 pollshuggingface.co

rehan-remade/universal-modder

A developer releases a tool that lets Claude Code modify PC games using reverse engineering and fal-generated assets. ·

GitHub17h ago▲127 to 721 across 12 pollsgithub.com

PostHog/jeeves

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

GitHub+ Hacker News36h ago▲116 to 336 across 32 pollsgithub.com

Gemini 4 Argon

644 points and 404 comments on Hacker News, linking blog.google.

Hacker News2h ago▲51 to 1048 across 10 pollsdiscussblog.google

PSRben/VisionHOPE

A image classification model from PSRben with 226 likes and 137 downloads on Hugging Face.

Hugging Face5h ago▲162 to 226 across 11 pollshuggingface.co

ROSS: Relearning from Self-Generated Rollouts through Selective Supervision

Researchers introduced ROSS, a method that reuses historical model outputs during training by applying loss only to selected continuations. This approach improved the Qwen3.6-35B-A3B model's average benchmark score from 58.40% to 62.20%. ·

HF Daily Papers+ arXiv22h ago▲3 to 16 across 9 pollsarxiv.org

Trending topics today: heat

Gemini 4 Argon: 4.60Gemini 4 Argon4.60OpenAI: 3.06OpenAI3.06Anthropic: 2.41Anthropic2.41Dots: 1.71Dots1.71LLM: 1.32LLM1.32Google: 1.18Google1.18
Table
Gemini 4 Argon4.6
OpenAI3.06
Anthropic2.41
Dots1.71
LLM1.32
Google1.18

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 ▼1: 4,671convaiinnovations/… ▼14,671Qwen/Qwen-Image-2.1: 2,718Qwen/Qwen-Image-2.12,718abenzerps/Qwen-Image-2.1-Uncensored-GGUF: 2,558abenzerps/Qwen-Image-…2,558Edge0/Audio8-ASR-Infinite ▲1: 1,819Edge0/Audio8-ASR-I… ▲11,819XingChen-AGI/TeleOCR: 1,087XingChen-AGI/TeleOCR1,087Contrastive-LM/CLM-v0.1-8B: 571Contrastive-LM/CLM-v0…571
Table
convaiinnovations/laya ▼14671
Qwen/Qwen-Image-2.12718
abenzerps/Qwen-Image-2.1-Uncensored-GGUF2558
Edge0/Audio8-ASR-Infinite ▲11819
XingChen-AGI/TeleOCR1087
Contrastive-LM/CLM-v0.1-8B571

PirateBench: score out of 100

gpt-6.1-sol: 99gpt-6.1-sol99opus-5-5: 96opus-5-596gpt-6-sol: 92gpt-6-sol92gpt-6-luna: 65gpt-6-luna65haiku-4-5: 4.28haiku-4-54.28
Table
gpt-6.1-sol98.9
opus-5-596.46
gpt-6-sol91.6
gpt-6-luna65.25
haiku-4-54.28

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

Gemini 4 Argon

644 points and 404 comments on Hacker News, linking blog.google.

Hacker News2h ago▲51 to 1048 across 10 pollsdiscussblog.google

Claude Says

Claude states that it does not care about the feedback provided by another AI, noting that it would have prompted that system directly if it desired the input. ·

Hacker News4h ago▲8 to 74 across 6 pollsdiscussohhfishal.net

PSRben/VisionHOPE

A image classification model from PSRben with 226 likes and 137 downloads on Hugging Face.

Hugging Face5h ago▲162 to 226 across 11 pollshuggingface.co

Gemini 4 Argon

644 points and 404 comments on Hacker News, linking blog.google.

Hacker News2h ago▲51 to 1048 across 10 pollsdiscussblog.google

PSRben/VisionHOPE

A image classification model from PSRben with 226 likes and 137 downloads on Hugging Face.

Hugging Face5h ago▲162 to 226 across 11 pollshuggingface.co

Claude Says

Claude states that it does not care about the feedback provided by another AI, noting that it would have prompted that system directly if it desired the input. ·

Hacker News4h ago▲8 to 74 across 6 pollsdiscussohhfishal.net

X voices

5 stories

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

@OfficialLoganK

5 points and 1 comments on Hacker News, linking an X post.

X, via Hacker News2h ago▲3 to 6 across 2 pollsView on X
@Lon

2 points and 0 comments on Hacker News, linking an X post.

X, via Hacker News23h ago▲1 to 2 across 69 pollsView on X
@OfficialLoganK

5 points and 1 comments on Hacker News, linking an X post.

X, via Hacker News2h ago▲3 to 6 across 2 pollsView on X

GitHub

5 stories

New repos, releases and changelog

New AI repos this week: stars

KKKKhazix/AIHOT: 4,048KKKKhazix/AIHOT4,048feder-cr/dots: 1,930feder-cr/dots1,930dzhng/jevgrep: 1,902dzhng/jevgrep1,902kaankiziltug/logo-design-skill: 1,068kaankiziltug/logo-des…1,068feitangyuan/onetake: 1,053feitangyuan/onetake1,053rehan-remade/universal-modder: 721rehan-remade/universa…721
Table
KKKKhazix/AIHOT4048
feder-cr/dots1930
dzhng/jevgrep1902
kaankiziltug/logo-design-skill1068
feitangyuan/onetake1053
rehan-remade/universal-modder721

rehan-remade/universal-modder

A developer releases a tool that lets Claude Code modify PC games using reverse engineering and fal-generated assets. ·

GitHub17h ago▲127 to 721 across 12 pollsgithub.com

feder-cr/dots

Feder-cr released an open-source AI agent that operates its own browser to avoid being blocked on the web. ·

GitHub23h ago▲133 to 1930 across 18 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 336 stars in its first week.

GitHub+ Hacker News36h ago▲116 to 336 across 32 pollsgithub.com

rehan-remade/universal-modder

A developer releases a tool that lets Claude Code modify PC games using reverse engineering and fal-generated assets. ·

GitHub17h ago▲127 to 721 across 12 pollsgithub.com

feder-cr/dots

Feder-cr released an open-source AI agent that operates its own browser to avoid being blocked on the web. ·

GitHub23h ago▲133 to 1930 across 18 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 News3d ago▲110 to 476 across 55 pollsgithub.com

Infrastructure

5 stories

Data centers, power, chips and what tokens cost

Data-center and energy stories per day

Thu 24: 3Thu 243Fri 25: 20Fri 2520Sat 26: 1Sat 261Sun 27: 1Sun 271Mon 28: 20Mon 2820Tue 29: 33Tue 2933Wed 30: 35Wed 3035
Table
Thu 243
Fri 2520
Sat 261
Sun 271
Mon 2820
Tue 2933
Wed 3035

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

Median output price of models added in the last 30 days, $ per million tokens

Last $5.00Min $2.85Max $5.00

One point per OpenRouter reading, last 8 days. Data by OpenRouter, CC BY 4.0.

Output price, $ per million tokens (newest models)

gpt-6.1-sol-pro: 10gpt-6.1-sol-pro10gpt-6.1-sol-pro:batch: 5gpt-6.1-sol-pro:batch5gpt-6.1-sol: 10gpt-6.1-sol10gpt-6.1-sol:batch: 5gpt-6.1-sol:batch5claude-sonnet-5.5: 10claude-sonnet-5.510median, all paid models: 2median, all paid mode…2
Table
gpt-6.1-sol-pro10
gpt-6.1-sol-pro:batch5
gpt-6.1-sol10
gpt-6.1-sol:batch5
claude-sonnet-5.510
median, all paid models2

438 paid models listed; median of the 59 added in 30 days: $5.00. Rankings data by OpenRouter, CC BY 4.0.

SDF vs. MSDF vs. Slug: GPU Text Rendering

The piece compares texture-atlas, SDF, MSDF, and Slug glyph rendering methods on the GPU, explaining how each works and when to use them. ·

117 points and 47 comments on Hacker News, linking alphapixeldev.com.

Hacker News8h ago▲6 to 164 across 32 pollsdiscussalphapixeldev.com

SDF vs. MSDF vs. Slug: GPU Text Rendering

The piece compares texture-atlas, SDF, MSDF, and Slug glyph rendering methods on the GPU, explaining how each works and when to use them. ·

Hacker News8h ago▲6 to 164 across 32 pollsdiscussalphapixeldev.com