Open swe uses deepagents under the hood Deepagents is general purpose, openswe is focused on coding
LLMS
-
Laurent Alexandre Warns Assembly: Claude Opus Surpasses Human Intelligence
By
–
Laurent Alexandre auditionné devant la Mission d'information sur l'IA à l'Assemblée.
— VISION IA (@vision_ia) 10 avril 2026
"Il faut sortir du déni et arrêter d'écouter Luc Julia."
"Claude Opus est beaucoup plus intelligent que moi, il m'écrabouille en médecine."
"Il y a vraiment le feu au lac et notre réflexion… https://t.co/aLjdUMhgCOLaurent Alexandre testifies before the AI Information Mission at the Assembly. "We need to snap out of denial and stop listening to Luc Julia." "Claude Opus is way smarter than me; it crushes me in medicine." "There's really a fire at the lake, and our thinking is way too
-

DMax: Aggressive Parallel Decoding for Distributed Language Models
By
–
DMax: Aggressive Parallel Decoding for dLLMs Paper: https://
huggingface.co/papers/2604.08
302
…
Code: https://
github.com/czg1225/DMax
Models: https://
huggingface.co/collections/Zi
geng/dmax-models
…
Datasets: https://
huggingface.co/collections/Zi
geng/dmax-training-data
… -
DMax: Aggressive Parallel Decoding for Diffusion Language Models
By
–
Wow, really fast!
— 机器之心 JIQIZHIXIN (@jiqizhixin) 10 avril 2026
Researchers at the National University of Singapore present DMax!
This new paradigm for diffusion language models (dLLMs) enables aggressive parallel decoding.
It cleverly mitigates error accumulation by reforming decoding as a progressive self-refinement… pic.twitter.com/T7ZS5jO4srWow, really fast! Researchers at the National University of Singapore present DMax! This new paradigm for diffusion language models (dLLMs) enables aggressive parallel decoding. It cleverly mitigates error accumulation by reforming decoding as a progressive self-refinement
-

Liquid AI Releases 34 Slides on Edge Model Design Training
By
–
I'm releasing the 34 slides on how we design and train best-in-class edge models at @liquidai I presented these slides yesterday at @aiDotEngineer They cover model architecture, pre-training, scaling laws, post-training, and even a solution to fix doom loops Special thanks to
-
AI Generates Website Content From Thousands of X Posts
By
–
My AI is reading everyone in AI here on X and it's amazing: https://
alignednews.com/ai It generated every word on that site after reading tens of thousands of posts here on X. -
The Evolution of Software Development From Code to Natural Language
By
–
The old model was simple:
you had an idea,
then translated it into syntax,
then fought the tooling until it worked. The new model looks very different: → describe the app in plain language
→ generate the interface, logic, and structure
→ test it
→ refine it through feedback -
From Syntax to Intent: The Future of Software Development
By
–
What if the biggest change in software isn’t a new language, but the fact that you may not need to think in code first anymore?
— Ronald van Loon (@Ronald_vanLoon) 10 avril 2026
I think we’re moving from:
syntax → intent
From:
debugging by trial and error → building through conversation
That changes who can build, how fast… pic.twitter.com/HTtXz4H0m3What if the biggest change in software isn’t a new language, but the fact that you may not need to think in code first anymore? I think we’re moving from: syntax → intent From: debugging by trial and error → building through conversation That changes who can build, how fast teams ship, and what developers actually spend time on. Here’s the breakdown, featuring what I learned from @GoogleAIStudio.
→ View original post on X — @ronald_vanloon, 2026-04-10 08:30 UTC
-
Apple Neural Engine’s Multimodal Power and CoreML Compiler Advances
By
–
I, too, underestimated the power of the Apple Neural Engine. AiDevCraft (@AiDevCraft) Going from text-only to multimodal in a single day while openly correcting benchmark numbers mid-thread is exactly the kind of rigorous iteration that makes edge ML credible. 99.78% ANE op mapping for a non-Apple architecture like Gemma 4 is the quietly impressive part — it means CoreML's compiler generalization is better than most people assume. — https://nitter.net/AiDevCraft/status/2042516832658297247#m
→ View original post on X — @scobleizer, 2026-04-10 08:18 UTC
-

LLM Routers Exploited: Malicious Tool Calls Stealing Credentials
By
–
Not even the tip of the iceberg. The Internet is becoming a wonderland for bad actors. Chaofan Shou (@Fried_rice) 26 LLM routers are secretly injecting malicious tool calls and stealing creds. One drained our client $500k wallet. We also managed to poison routers to forward traffic to us. Within several hours, we can directly take over ~400 hosts. Check our paper: arxiv.org/abs/2604.08407 — https://nitter.net/Fried_rice/status/2042423713019412941#m
→ View original post on X — @scobleizer, 2026-04-10 07:58 UTC