SupersonicLabs/Julia-1
A text classification model from SupersonicLabs with 135 likes and 274 downloads on Hugging Face.
Top now
6 storiesWhat is climbing fastest across every source
Image-2.17.0xQwen3.87.0xLLMs6.7xQwen7.0xV2.67.0xXiaomiMiMo7.0xDeepSeek7.0xOpenAI6.6xQwen-Image-2.17.0xconvaiinnovations7.0x
"As a Language Model": Chat Template Switches LLM Self-Referential Voice
A study found that chat templates cause large language models to use self-referential disclaimers. This effect was observed across several open-source models. ·
AI:AM: What If It Works Too Well? Colluding Agents, $200M Safety Orgs, Virtual Cells Saturate at 2%
A podcast episode where hosts interview five experts about risks in multi-agent coordination, safety funding limits, and compute markets. The discussion also covers sensor-based models and robotic drug testing. ·
SupraLabs/Supra2-IMG
A text to image model from SupraLabs with 140 likes and 515 downloads on Hugging Face.
OpenAI Feared "Optics" of what might appear on Hacker News
416 points and 337 comments on Hacker News, linking authorsguild.org.
SupersonicLabs/Julia-1
A text classification model from SupersonicLabs with 135 likes and 274 downloads on Hugging Face.
"As a Language Model": Chat Template Switches LLM Self-Referential Voice
A study found that chat templates cause large language models to use self-referential disclaimers. This effect was observed across several open-source models. ·
OpenAI Feared "Optics" of what might appear on Hacker News
416 points and 337 comments on Hacker News, linking authorsguild.org.
Edge0/Audio8-ASR-Infinite
A automatic speech recognition model from Edge0 with 969 likes and 19.4k downloads on Hugging Face.
unslothai/unsloth Flash-Attention2, Causal-Conv1D, Mamba_SSM Binaries
Prebuilt Linux x86_64 wheels for flash-attn, causal-conv1d, and mamba-ssm are available. They support CUDA 13, PyTorch 2.13 and 2.14, and Python 3.13. ·
lemomo-ai/lemo-opuscar
A repository offers thirty-nine film styles, each with a reusable prompt and a short code-generated sample. Users can select a style and use an agent to direct their own story. ·
"As a Language Model": Chat Template Switches LLM Self-Referential Voice
A study found that chat templates cause large language models to use self-referential disclaimers. This effect was observed across several open-source models. ·
69 points and 67 comments on Hacker News, linking arxiv.org.
OpenAI Feared "Optics" of what might appear on Hacker News
416 points and 337 comments on Hacker News, linking authorsguild.org.
Edge0/Audio8-ASR-Infinite
A automatic speech recognition model from Edge0 with 969 likes and 19.4k downloads on Hugging Face.
StarDoc-AI/TeleOCR
A image text to text model from StarDoc-AI with 536 likes and 27.8k downloads on Hugging Face.
unslothai/unsloth Flash-Attention2, Causal-Conv1D, Mamba_SSM Binaries
Prebuilt Linux x86_64 wheels for flash-attn, causal-conv1d, and mamba-ssm are available. They support CUDA 13, PyTorch 2.13 and 2.14, and Python 3.13. ·
lemomo-ai/lemo-opuscar
A repository offers thirty-nine film styles, each with a reusable prompt and a short code-generated sample. Users can select a style and use an agent to direct their own story. ·
Trending topics today: heat
Table
| Image-2.1 | 2.42 |
| Qwen3.8 | 1.73 |
| LLMs | 1.62 |
| Qwen | 1.42 |
| V2.6 | 1.02 |
| XiaomiMiMo | 1.02 |
Heat = summed hotness of today's stories naming the topic.
Models
5 storiesReleases, papers, trending weights and new listings
Top 6 trending on Hugging Face: likes
Table
| convaiinnovations/laya | 4008 |
| Qwen/Qwen-Image-2.1 | 2448 |
| abenzerps/Qwen-Image-2.1-Uncensored-GGUF | 2017 |
| XingChen-AGI/Xing4.0-29B-A4B | 1774 |
| Edge0/Audio8-ASR-Infinite | 969 |
| Altworld/Hemmingway-1 | 724 |
StarDoc-AI/TeleOCR
A image text to text model from StarDoc-AI with 536 likes and 27.8k downloads on Hugging Face.
abenzerps/Qwen-Image-2.1-Uncensored-GGUF
A text to image model from abenzerps with 2,017 likes and 964.2k downloads on Hugging Face.
convaiinnovations/laya
A text classification model from convaiinnovations with 4,008 likes and 0 downloads on Hugging Face.
Contrastive-LM/CLM-v0.1-8B
A text ranking model from Contrastive-LM with 362 likes and 766 downloads on Hugging Face.
Lightricks/LTX-2.5
A image to video model from Lightricks with 5,285 likes and 1.6M downloads on Hugging Face.
Edge0/Audio8-ASR-Infinite
A automatic speech recognition model from Edge0 with 969 likes and 19.4k downloads on Hugging Face.
StarDoc-AI/TeleOCR
A image text to text model from StarDoc-AI with 536 likes and 27.8k downloads on Hugging Face.
convaiinnovations/laya
A text classification model from convaiinnovations with 4,008 likes and 0 downloads on Hugging Face.
TaichuAI/ZDTaichu5.0-9B
A image text to text model from TaichuAI with 1,677 likes and 11.6k downloads on Hugging Face.
Contrastive-LM/CLM-v0.1-8B
A text ranking model from Contrastive-LM with 362 likes and 766 downloads on Hugging Face.
Edge0/Audio8-ASR-Infinite
A automatic speech recognition model from Edge0 with 969 likes and 19.4k downloads on Hugging Face.
StarDoc-AI/TeleOCR
A image text to text model from StarDoc-AI with 536 likes and 27.8k downloads on Hugging Face.
convaiinnovations/laya
A text classification model from convaiinnovations with 4,008 likes and 0 downloads on Hugging Face.
TaichuAI/ZDTaichu5.0-9B
A image text to text model from TaichuAI with 1,677 likes and 11.6k downloads on Hugging Face.
Contrastive-LM/CLM-v0.1-8B
A text ranking model from Contrastive-LM with 362 likes and 766 downloads on Hugging Face.
X voices
5 storiesX posts other sources are linking to, then named voices on Bluesky and Mastodon
@mitchellh
An agent in a loop optimizing a renderer with the goal to minimize frame times
1 points and 0 comments on Hacker News, linking an X post.
@amplifiedamp
Font where each token is equal-width
1 points and 0 comments on Hacker News, linking an X post.
@mitchellh
An agent in a loop optimizing a renderer with the goal to minimize frame times
1 points and 0 comments on Hacker News, linking an X post.
@amplifiedamp
Font where each token is equal-width
1 points and 0 comments on Hacker News, linking an X post.
@benwills
"I know I am GPT-6. I cannot confirm whether this session uses GPT-6-Sol."
4 points and 2 comments on Hacker News, linking an X post.
@claudeai
A thread of early explorations with Claude Opus 5.5
1 points and 0 comments on Hacker News, linking an X post.
GitHub
5 storiesNew repos, releases and changelog
New AI repos this week: stars
Table
| jev-chat/jev-chat-jarvis | 6747 |
| unreallabsai/unreal-agent | 1979 |
| yetone/magpie | 1182 |
| mikehasa/golive-skill | 996 |
| freestylefly/WeChatBridge | 832 |
| nokia-applied-research/AnyJev | 809 |
unslothai/unsloth Flash-Attention2, Causal-Conv1D, Mamba_SSM Binaries
Prebuilt Linux x86_64 wheels for flash-attn, causal-conv1d, and mamba-ssm are available. They support CUDA 13, PyTorch 2.13 and 2.14, and Python 3.13. ·
New release prebuilt-wheels-cu13 of unslothai/unsloth on GitHub.
unslothai/unsloth Flash-Attention2, Causal-Conv1D, Mamba_SSM Binaries
Prebuilt Linux x86_64 wheels for flash-attn, causal-conv1d, and mamba-ssm are available. They support CUDA 13, PyTorch 2.13 and 2.14, and Python 3.13. ·
New release prebuilt-wheels-cu13 of unslothai/unsloth on GitHub.
lemomo-ai/lemo-opuscar
A repository offers thirty-nine film styles, each with a reusable prompt and a short code-generated sample. Users can select a style and use an agent to direct their own story. ·
feitangyuan/onetake
This is a Claude Agent Skill for creating continuous motion films for product launches and feature demos, where each segment flows seamlessly from the previous one. ·
dzhng/jevgrep
This is a command-line tool for coding agents. It uses Jev to locate relevant files and source code by describing functionality. ·
openai/codex 0.157.1
OpenAI released version 0.157.1 of its Codex tool. No specific feature changes were identified because the provided release notes were empty and the comparison link was unavailable. ·
unslothai/unsloth Flash-Attention2, Causal-Conv1D, Mamba_SSM Binaries
Prebuilt Linux x86_64 wheels for flash-attn, causal-conv1d, and mamba-ssm are available. They support CUDA 13, PyTorch 2.13 and 2.14, and Python 3.13. ·
New release prebuilt-wheels-cu13 of unslothai/unsloth on GitHub.
lemomo-ai/lemo-opuscar
A repository offers thirty-nine film styles, each with a reusable prompt and a short code-generated sample. Users can select a style and use an agent to direct their own story. ·
feitangyuan/onetake
This is a Claude Agent Skill for creating continuous motion films for product launches and feature demos, where each segment flows seamlessly from the previous one. ·
dzhng/jevgrep
This is a command-line tool for coding agents. It uses Jev to locate relevant files and source code by describing functionality. ·
anthropics/claude-code v2.1.283
Claude Code v2.1.283 adds gateway headers, model management settings, and OpenTelemetry logging. It also introduces prompt auditing, expanded error details, and a load testing mode. ·
Data centers & energy
5 storiesBuilds, power deals and the grid
Data-center and energy stories per day
Table
| Mon 21 | 2 |
| Tue 22 | 3 |
| Wed 23 | 3 |
| Thu 24 | 2 |
| Fri 25 | 20 |
| Sat 26 | 1 |
| Sun 27 | 1 |
Stories from DCD, Data Center Knowledge, EIA and Hacker News, by publication day.
AI infrastructure will define the next industrial revolution
The article argues that artificial intelligence infrastructure will shape the next industrial revolution. It notes that energy, intelligence, and digital systems are delivering widespread global benefits. ·
From DCD.
Efforts to Curb Data Center Speculation Gain Ground Across the US
US states are implementing measures like large-load tariffs to reduce speculative data center projects amid growing public opposition. ·
Google’s Grid-Interactive AI Data Centers: From Backup to Grid Partner
Google presented a strategy for AI data centers to act as grid partners. The approach uses full-stack design, low-voltage direct current, battery storage, and liquid cooling to support local power networks. ·
Vocus to build 4,000km fiber link between Brisbane and Darwin
Vocus is constructing a 4,000-kilometer fiber optic connection linking Brisbane and Darwin. The project targets rising demand from hyperscalers, neoclouds, and enterprise clients. ·
AI infrastructure will define the next industrial revolution
The article argues that artificial intelligence infrastructure will shape the next industrial revolution. It notes that energy, intelligence, and digital systems are delivering widespread global benefits. ·
From DCD.
Efforts to Curb Data Center Speculation Gain Ground Across the US
US states are implementing measures like large-load tariffs to reduce speculative data center projects amid growing public opposition. ·
Google’s Grid-Interactive AI Data Centers: From Backup to Grid Partner
Google presented a strategy for AI data centers to act as grid partners. The approach uses full-stack design, low-voltage direct current, battery storage, and liquid cooling to support local power networks. ·
Vocus to build 4,000km fiber link between Brisbane and Darwin
Vocus is constructing a 4,000-kilometer fiber optic connection linking Brisbane and Darwin. The project targets rising demand from hyperscalers, neoclouds, and enterprise clients. ·
Hardware prices
2 storiesGPUs, memory, and what tokens cost
Output price, $ per million tokens (newest models)
Table
| perceptron-mk1.5 | 1.5 |
| ember-1 | 15 |
| glm-5.3-prime | 8.8 |
| qwen3.8-max-prime | 12 |
| aion-3.5-mini | 1.4 |
| median, all paid models | 2 |
431 paid models listed; median of the 52 added in 30 days: $4.00. Rankings data by OpenRouter, CC BY 4.0.
Virtio-nvgpu: Near-native Nvidia GPU access inside a KVM guest
An experimental project provides a virtio device that allows KVM virtual machines to access NVIDIA GPUs with near-native performance. ·
154 points and 65 comments on Hacker News, linking github.com.
Raspberry Pi blocks changing RAM chips
258 points and 211 comments on Hacker News, linking forums.raspberrypi.com.