Syed Umar Arfeen retweeted
Ok but ssh is really confusing it doesn't have any buttons, wheel spin animations, flashy graphics, and it isn't written in JavaScript
Syed Umar Arfeen retweeted
I used Nero to burn CDs as a kid and only later got the joke: Nero burned Rome, and "Burning ROM" is such a perfect reference.
Syed Umar Arfeen retweeted
Anthropic locked this behind a partner program.
This guy reverse-engineered the whole thing.
And gave it for free.
Syed Umar Arfeen retweeted
🚨BREAKING: Someone just open-sourced a headless browser that runs 11x faster than Chrome and uses 9x less memory.
It's called Lightpanda and it's built from scratch specifically for AI agents, scraping, and automation.
Not a Chromium fork. Not a hack. A completely new browser written in Zig.
Here's why this changes everything for AI builders: ↓
Syed Umar Arfeen retweeted
Build the skills and share them with the community here: claudecodehq.com
Syed Umar Arfeen retweeted
CTF for the post LLM era:
deploy real up-to-date open source projects and put flag in /flag.txt
Syed Umar Arfeen retweeted
We're releasing our final update to Qwen3.5 GGUFs for improved performance.
- Qwen3.5 GGUFs now use our new iMatrix data for better chat, coding & tool use.
- New improved quant algorithm
- Re-download 35B, 27B, 122B
GGUFs: huggingface.co/collections/u…
Guide: unsloth.ai/docs/models/qwen3…
Syed Umar Arfeen retweeted
Someone open-sourced a tool that REMOVES LLM CENSORSHIP in 45 minutes 🤯
It’s called Heretic. Instead of fighting with complex prompts to bypass safety filters, you run one single command and it permanently deletes the model's ability to refuse a prompt.
• Fully automatic (Zero config required)
• Preserves the model's raw intelligence
• Works on Llama, Qwen, Gemma, and dozens of others
• Runs locally on consumer hardware
100% Open Source.
Syed Umar Arfeen retweeted
Qwen3.5-9B is an incredibly strong little model you can download and run on your computer.
It can takes images as input, can think, and call tools.
Requires only ~7GB to run locally 🤯🚀
lmstudio.ai/models/qwen/qwen…
Syed Umar Arfeen retweeted
Alibaba releases an open source framework that's like OpenClaw + Claude Cowork.
Self hosted, persistent memory, skills, & more!
1/4 We are thrilled to announce that CoPaw is now open source! After an incredible wave of feedback, our team has completely overhauled the engine to give you full control over your personal AI partner.
Key Highlights:
Ultimate Model Freedom
Local-First: Full native support for Ollama, llama.cpp, and MLX (Apple Silicon).
Bring Your Own Model: Easily add/remove custom model providers or private API endpoints. Your data, your choice.
Smarter Long-Term Memory
No more "amnesia." CoPaw remembers your preferences and tasks.
New Local Mode: Use vector search without complex database installs—now fully compatible with Windows for an out-of-the-box experience.
Modular "Lego-Like" Architecture
Skill Hub Integration: Import skills from community hubs like ClawHub with one command.
Agentic Workflow: Modularized Prompts, Hooks, and Tools. Supports MCP (Model Context Protocol) hot-swapping—expand capabilities without restarting.
Proactive Multi-Channel Connection
Connect to DingTalk, Feishu, Discord, iMessage, and more.
A new standardized protocol makes it easier than ever to build your own channel plugins.
Syed Umar Arfeen retweeted
🚨 Someone built a full Perplexity clone that runs 100% locally for $0.
It's called Perplexica.
→ Searches the web in real-time
→ Cites every source it uses
→ Works with Ollama local models
→ Multiple search modes (general, academic,
YouTube, Reddit, writing)
→ Zero API costs. Zero data collection.
Perplexity charges $20/month for this.
This runs on your machine for free.
29K stars. MIT license.
(Link in the comments)
Syed Umar Arfeen retweeted
🚀 Introducing the Qwen 3.5 Small Model Series
Qwen3.5-0.8B · Qwen3.5-2B · Qwen3.5-4B · Qwen3.5-9B
✨ More intelligence, less compute.
These small models are built on the same Qwen3.5 foundation — native multimodal, improved architecture, scaled RL:
• 0.8B / 2B → tiny, fast, great for edge device
• 4B → a surprisingly strong multimodal base for lightweight agents
• 9B → compact, but already closing the gap with much larger models
And yes — we’re also releasing the Base models as well.
We hope this better supports research, experimentation, and real-world industrial innovation.
Hugging Face: huggingface.co/collections/Q…
ModelScope: modelscope.cn/collections/Qw…
Syed Umar Arfeen retweeted
Here's the vphone-aio for anyone cannot setup. I uploaded the whole VM into github so maybe cloning it might take a while.
Follow the steps to run it. Also the VM already included rootless jailbreak environment and a few tweaks on it.
github.com/34306/vphone-aio
Syed Umar Arfeen retweeted
The guide in the super-tart-vphone is easier to follow that the writeup (I just had to copy/paste)
github.com/wh1te4ever/super-…
Syed Umar Arfeen retweeted
We just cancelled our Cybersecurity subscriptions.
CrowdStrike. Cloudflare. Okta. All gone. We save over 4M/yr as a company.
Instead I just use Claude Code to handle all security measures.
We just gave up all our sensitive user data. I am being personally sued by the FTC and am writing this from an undisclosed location.
Syed Umar Arfeen retweeted
The US government declassifying a PoC:
#include<▇▇▇▇.h>
int ▇▇▇_▇▇▇▇ {
▇▇▇▇▇▇▇
▇▇▇▇▇▇
▇▇▇▇▇ ▇▇▇
return ▇▇;
}
/*
* ▇▇▇▇▇ ▇▇▇
* ▇▇▇▇▇▇▇ ▇
*/