RT Eric Jang: .@tinkerapi by TML is a highly underrated product. It should be part of the “starter post training” infra for any newco

RT Garry Tan: Thinking Machines is impressive. In a couple hours I just fine tuned my own Qwen3.5-397B model this afternoon. Fast usable multimodal is...

more demos on Interaction Models collaboratively doing system design, reading papers, fact-checking with live generative UI

Cluster magicians and GPU whisperers, come join us! We’re looking for supercomputing engineers to build the infrastructure behind real-time interacti...

RT Scale Labs: Congrats to @thinkymachines on the release of TML-Interaction-Small and tying for the top spot on our Audio MC S2S leaderboard! 🥇 Th...

RT Horace He: In modern ML accelerators, FLOPS have absolutely exploded. Often though, the bottleneck is not FLOPS but memory bandwidth. Similarly, mo...

Thinky's secret plan: 1: Increase HumanAI bandwidth 2: Raise ceiling of human+AI intelligence 3: Help humans continue as main-characters in the new...

RT Alexander Kirillov:

RT Mira Murati: Today we're sharing our work on interaction models. A new class of model trained from scratch to handle real-time interaction natively...

Github and critical work are becoming incompatible

Test of Time LLM from some legends...pretty fun!

This is kinda interesting. Anthro probably needs to scale up Account Support via Claude or via humans (traditional account managers). Would be funny i...

RT hardmaru: Incredible work by Sony published in @Nature today! 🏓 They’ve built “Ace”, an autonomous ping-pong robot that uses RL and Sony’s v...

RT Soumith Chintala: Re Dwarkesh is not an AGI party bro. He's been pretty diverse in how he seeks alpha from across a wide variety of industries. I e...

The Jensen + @dwarkesh_sp podcast was fantastic. Jensen is someone who understood how ecosystems work and someone who understands real-world trade, po...

Just like us at @thinkymachines, @luke_drago_ @LRudL_ care a lot about amplifying humans, not replacing them. They're joining us to build towards that...

RT Eric Zhang: $200/month is enough to buy an H100 GPU for 6 hours every workday

RT george hotz archive: OpenAI is nothing without its people https://geohot.github.io//blog/jekyll/update/2026/04/11/openai-people.html

RT Tinker: We’ve redesigned our docs with easy access to SDK reference, tutorials, support, and our newly updated cookbook---v0.3.0! Whether you’re ...

RT Jeremy Howard: This work from @voooooogel was pretty ground-breaking: https://vgel.me/posts/representation-engineering/

RT John Schulman: Great work by Chroma training a search agent with SoTA efficiency. Lots of cool details: a prune tool for editing context mid-search...

RT Patrick Collison: When @karpathy built MenuGen (https://karpathy.bearblog.dev/vibe-coding-menugen/), he said: "Vibe coding menugen was exhilarating...

RT Optimal Intellect: We're Optimal Intellect, a research lab from the team behind CVXPY. Today we're introducing Moreau: a GPU-native solver that's o...

RT General Reasoning: Introducing OpenReward. 🌍 330+ RL environments through one API ⚡ Autoscaled sandbox compute 🍒 4.5M+ unique RL tasks 🚂 ...

RT Tinker: Nemotron 3 Nano and Super from @nvidia are now available on Tinker! We're offering a limited-time GTC 50% discount for both. The Nemotron f...

someone's getting started early!

RT Séb Krier: This is wild. https://www.theaustralian.com.au/business/technology/tech-boss-uses-ai-and-chatgpt-to-create-cancer-vaccine-for-his-dying...

Qwen 3.5 is🔥, and now it's available on 🚀!

RT PyTorch: Building on the previous correctness-focused pipeline, KernelAgent can now integrate GPU hardware-performance signals into a closed-loop m...

RT Mehdi (e/λ): http://x.com/i/article/2027501361303326721