Joseph Reth

CEO @autopoiesislab

[François Chollet @fchollet]

If you're not obsessed with the research problem you're working on, for its own sake, you're unlikely to succeed. Intrinsic motivation is far more powerful than external rewards.

[Michael Belhassen @michaelblhssn]

Meet Anoria.
The first wearable that reads your emotions so you can enhance your EQ.

Welcome to the EQ era.

RT + comment "EQ" to receive an invitation to pre-order.

[Kimi.ai @Kimi_Moonshot]

Meet Kimi K2.6: Advancing Open-Source Coding

🔹Open-source SOTA on HLE w/ tools (54.0), SWE-Bench Pro (58.6), SWE-bench Multilingual (76.7), BrowseComp (83.2), Toolathlon (50.0), Charxiv w/ python(86.7), Math Vision w/ python (93.2)

What's new:
🔹Long-horizon coding - 4,000+ tool calls, over 12 hours of continuous execution, with generalization across languages (Rust, Go, Python) and tasks (frontend, devops, perf optimization).
🔹Motion-rich frontend - Videos in hero sections, WebGL shaders, GSAP + Framer Motion, Three.js 3D.
🔹Agent Swarms, elevated - 300 parallel sub-agents × 4,000 steps per run (up from K2.5's 100 / 1,500). One prompt, 100+ files.
🔹Proactive Agents - K2.6 model powers OpenClaw, Hermes Agent, etc for 24/7 autonomous ops.
🔹Claw Groups (research preview) - bring your own agents, command your friends', bots & humans in the loop.
-
K2.6 is now live on https://t.co/YutVbwktG0 in chat mode and agent mode.
For production-grade coding, pair K2.6 with Kimi Code: https://t.co/uvoSJKyGCY
-
🔗 API: https://t.co/EOZkbOwCN4
🔗 Tech blog: https://t.co/9wWvgIQSS3
🔗 Weights & code: https://t.co/Be0hjs2RTP

[Chubby♨️ @kimmonismus]

Free AI scientists for the US: Aristotle, a next-generation AI co-scientist, is launching its X1 family and Instant models, built in closed beta with researchers from Harvard, Stanford, and the NIH and is now completely free for verified U.S. scientists.

The platform tackles real research workflows, from literature synthesis and hypothesis generation to experimental validation, with models like X1 Verify (self-critiquing reasoning), X1 Search (knowledge-graph-powered discovery), and X1 Spark (ranked bold hypotheses).

[AshutoshShrivastava @ai_for_success]

We will make more scientific discoveries in the next decade with the help of AI than in the last 100 years, for sure.

Imagine how much scientists could do if they could go through thousands of papers, find the gaps, and test hypotheses faster with the help of AI. The future is freaking amazing.

Autopoiesis has launched Aristotle, an AI co scientist, and it is free for verified researchers in the US.

[Autopoiesis @autopoiesislab]

It's here. Aristotle is now live.

AI built for how scientists actually think. Self-skeptical reasoning, epistemic graph exploration, bold hypotheses grounded after generation.

Free for verified researchers in the United States.

https://t.co/SOyuXL4IUq

[Gary Simon @designcoursecom]

Only OG's will remember.

[Jeremy Howard @jeremyphoward]

til some folks nowadays consider running a normal linux web server like we've all done for decades to be a "non-standard stack"

feels like some kind of learned helplessness or something

[Sam Lambert @samlambert]

Until ~2015, GitHub Pages hosted over 2 million websites on 2 servers with a multi-million-line nginx.conf, edited and reloaded per deploy. This worked incredibly well, with https://t.co/DcP1J23VVj ranking as the 140th most visited domain on the web at the time.

[ATOM TEAM @the_atomgame]

Just a regular NPC dialogue tree in our game:

[François Chollet @fchollet]

The most important skill for a researcher is not technical ability. It's taste. The ability to identify interesting and tractable problems, and recognize important ideas when they show up.

This can't be taught directly. It's cultivated through curiosity and broad reading.

[François Chollet @fchollet]

The most elegant solutions look inevitable in retrospect. They aren't some complex chain of tricks you've engineered out of nothing, but rather a simple abstraction that faithfully mirrors the shape of the problem.

[WarrenBuffering @WarrenInTheBuff]

An experienced af dev built this

Much easier than trying to build something circular with nearly identical utility

Proof: Nobody knew

[Quant Beckman @quantbeckman]

This is the reason why Renaissance hires PhDs to clean data 😬

[Autopoiesis @autopoiesislab]

We've selected @Oracle Cloud Infrastructure to power the next phase of our AI co-scientist development.

Scaling with enterprise-grade AI infrastructure. Building AI systems that researchers can trust.

https://t.co/o7XwlhCBwp

[Qwen @Alibaba_Qwen]

🚀 Qwen3-30B-A3B-2507 and Qwen3-235B-A22B-2507 now support ultra-long context—up to 1 million tokens!

🔧 Powered by:

• Dual Chunk Attention (DCA) – A length extrapolation method that splits long sequences into manageable chunks while preserving global coherence. 

• MInference – Sparse attention that cuts overhead by focusing on key token interactions

💡 These innovations boost both generation quality and inference speed, delivering up to 3× faster performance on near-1M token sequences.
✅ Fully compatible with vLLM and SGLang for efficient deployment.

📄 See the update model cards for how to enable this feature.
https://t.co/WufQYUhImD
https://t.co/9eg5iAut4d
https://t.co/OglPSNk8lz
https://t.co/G0JNw0XZGF
https://t.co/KOLTlDujFE
https://t.co/wXC3NLztep
https://t.co/PwWUW02Pgd
https://t.co/q1MsrwgjQz

[The AI Colony @TheAIColony]

🚨BREAKING: The world’s first self-verifying AI scientist just launched and it’s INSANE!

Aristotle X1 Verify doesn’t just answer, it peer-reviews its reasoning before speaking.

SPOILER: OpenAI is now behind.

Everything you need to know:

[mel @melqtx]

what killed the hacker culture

[The AI Colony @TheAIColony]

🚨Breaking: Aristotle X1 just scored 92.4% on GPQA (a brutal science test for AIs) and 96.1% on SimpleQA.

It’s not from OpenAI. Not from Anthropic.
It’s from a startup called Autopoiesis.

This thing might be smarter than your smartest friend.

[Healthcare AI Guy @HealthcareAIGuy]

NEW: Autopoiesis, building the OpenEvidence platform for science, just raised funding led by Informed Ventures.

Its AI model ‘Aristotle’ hit 92.4% on GPQA Diamond & 96.1% on SimpleQA -- beating OpenAI & Google with a self-skeptical architecture for trustworthy science.