Archived front page, Wed 30 Sep 2026, 06:00 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, 06:00 UTC

Top now

6 stories

What is climbing fastest across every source

OpenAI6.3xAnthropic6.4xGLM-5.37.0xVideo Generation6.8xGPT-6.17.0xMuse7.4xClaude7.2xLanguage6.7xLearning5.2xDevDay7.1x

MaLiang-Harness: A Programmable Path to Image and Video Generation

Researchers introduced MaLiang-Harness, a framework that uses multimodal models to iteratively build, check, and fix code for image and video generation. The system evaluates eleven closed-source models on new benchmarks to measure success and quality. ·

HF Daily Papers6h ago▲ 55 to 87 across 4 pollsarxiv.org

feder-cr/dots

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

GitHub7h ago▲ 3133 to 275 across 2 pollsgithub.com

MaLiang-Harness: A Programmable Path to Image and Video Generation

Researchers introduced MaLiang-Harness, a framework that uses multimodal models to iteratively build, check, and fix code for image and video generation. The system evaluates eleven closed-source models on new benchmarks to measure success and quality. ·

HF Daily Papers6h ago▲ 785 to 87 across 4 pollsarxiv.org

feder-cr/dots

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

GitHub7h ago▲ 75133 to 275 across 2 pollsgithub.com

openai/codex 0.159.2

OpenAI released version 0.159.2 of Codex, a patch that stops console windows from flashing on Windows when the tool runs background processes. ·

GitHub Releases6h ago▲ 690 to 1 across 3 pollsgithub.com

PostHog/jeeves

233 points and 89 comments on Hacker News, linking github.com.

GitHub+ Hacker News20h ago▲ 154116 to 301 across 16 pollsgithub.com

MaLiang-Harness: A Programmable Path to Image and Video Generation

Researchers introduced MaLiang-Harness, a framework that uses multimodal models to iteratively build, check, and fix code for image and video generation. The system evaluates eleven closed-source models on new benchmarks to measure success and quality. ·

HF Daily Papers6h ago▲ 925 to 87 across 4 pollsarxiv.org

JoasASantos/Offensive-Security-AI-Models

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

GitHub+ Hacker News2d ago▲ 92110 to 419 across 39 pollsgithub.com

Trending topics today: heat

OpenAI: 3.83OpenAI3.83Anthropic: 2.60Anthropic2.60GLM-5.3: 1.49GLM-5.31.49Video Generation: 1.31Video Generation1.31GPT-6.1: 1.11GPT-6.11.11Muse: 1.06Muse1.06
Table
OpenAI3.83
Anthropic2.6
GLM-5.31.49
Video Generation1.31
GPT-6.11.11
Muse1.06

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: 4,534convaiinnovations/laya4,534Qwen/Qwen-Image-2.1 ▼1: 2,669Qwen/Qwen-Image-2.1 ▼12,669abenzerps/Qwen-Image-2.1-Uncensored-GGUF ▼1: 2,459abenzerps/Qwen-Ima… ▼12,459Edge0/Audio8-ASR-Infinite ▲1: 1,552Edge0/Audio8-ASR-I… ▲11,552XingChen-AGI/TeleOCR ▲1: 884XingChen-AGI/TeleO… ▲1884Contrastive-LM/CLM-v0.1-8B ▲1: 532Contrastive-LM/CLM… ▲1532
Table
convaiinnovations/laya4534
Qwen/Qwen-Image-2.1 ▼12669
abenzerps/Qwen-Image-2.1-Uncensored-GGUF ▼12459
Edge0/Audio8-ASR-Infinite ▲11552
XingChen-AGI/TeleOCR ▲1884
Contrastive-LM/CLM-v0.1-8B ▲1532

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/

VoxMem: Benchmarking Multimodal Memory in Large Audio Language Models

Researchers introduced VoxMem, a benchmark containing 3,196 evaluation instances across 177 hours of audio. It tests how large audio language models retain and process multimodal memory from spoken conversations. ·

19 upvotes on Hugging Face Daily Papers.

HF Daily Papers6h ago▲ 5arxiv.org

Think Before You Score: Thinking Reward Model for Visual Generation

Researchers introduced a thinking reward model that creates case-specific evaluation criteria before scoring visual outputs. They also developed a new optimization method to improve reward discrimination while maintaining detailed pointwise scoring. ·

HF Daily Papers6h ago▲ 58 to 29 across 3 pollsarxiv.org

Think Before You Score: Thinking Reward Model for Visual Generation

Researchers introduced a thinking reward model that creates case-specific evaluation criteria before scoring visual outputs. They also developed a new optimization method to improve reward discrimination while maintaining detailed pointwise scoring. ·

HF Daily Papers6h ago▲ 758 to 29 across 3 pollsarxiv.org

Language models for text classification: From bag-of-words to Jev

The source text lists titles for visual guides and experiments on recurrent neural networks, convolutional neural networks, and transformers, focusing on their accuracy and efficiency. ·

69 points and 2 comments on Hacker News, linking magazine.sebastianraschka.com.

Hacker News+ Ahead of AI19h ago▲ 1029 to 71 across 27 pollsdiscussmagazine.sebastianraschka.com

Claude Sonnet 5.5

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

Hacker News+ Simon Willison36h ago▲ 9465 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 News2d ago▲ 92110 to 419 across 39 pollsgithub.com

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+ arXiv6h ago▲ 903 to 5 across 3 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: 3,512KKKKhazix/AIHOT3,512dzhng/jevgrep: 1,803dzhng/jevgrep1,803feitangyuan/onetake: 992feitangyuan/onetake992kaankiziltug/logo-design-skill: 886kaankiziltug/logo-des…886lemomo-ai/lemo-opuscar: 623lemomo-ai/lemo-opuscar623glanderness/BeefTV: 565glanderness/BeefTV565
Table
KKKKhazix/AIHOT3512
dzhng/jevgrep1803
feitangyuan/onetake992
kaankiziltug/logo-design-skill886
lemomo-ai/lemo-opuscar623
glanderness/BeefTV565

google-gemini/gemini-cli Release v0.62.0

Google released version 0.62.0 of the Gemini CLI. The update includes fixes for OAuth token handling, UI layout rendering, and MCP tool call formatting. ·

GitHub Releases9h ago▲ 0github.com

GPT-6.1 Sol in GitHub Copilot

OpenAI's GPT-6.1 Sol model is now generally available in GitHub Copilot for agentic coding and terminal workflows. ·

GitHub Changelog13h ago▲ 0github.blog

openai/codex 0.159.2

OpenAI released version 0.159.2 of Codex, a patch that stops console windows from flashing on Windows when the tool runs background processes. ·

GitHub Releases6h ago▲ 690 to 1 across 3 pollsgithub.com

openai/codex 0.159.1

OpenAI updated Codex version 0.159.1 to make GPT-6.1 Sol the default model in its bundled and Amazon Bedrock catalogs. ·

GitHub Releases9h ago▲ 640 to 7 across 5 pollsgithub.com

PostHog/jeeves

233 points and 89 comments on Hacker News, linking github.com.

GitHub+ Hacker News20h ago▲ 154116 to 301 across 16 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 News2d ago▲ 92110 to 419 across 39 pollsgithub.com

openai/codex 0.159.1

OpenAI updated Codex version 0.159.1 to make GPT-6.1 Sol the default model in its bundled and Amazon Bedrock catalogs. ·

GitHub Releases9h ago▲ 840 to 7 across 5 pollsgithub.com

openai/codex 0.159.2

OpenAI released version 0.159.2 of Codex, a patch that stops console windows from flashing on Windows when the tool runs background processes. ·

GitHub Releases6h ago▲ 790 to 1 across 3 pollsgithub.com

Data centers & energy

5 stories

Builds, power deals and the grid

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: 2Wed 302
Table
Thu 243
Fri 2520
Sat 261
Sun 271
Mon 2820
Tue 2933
Wed 302

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

Hardware prices

2 stories

GPUs, memory, and what tokens cost

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 60 added in 30 days: $5.00. Rankings data by OpenRouter, CC BY 4.0.