LonelyGuy ./

Introducing "Guess What?" an AI-orchestrated interactive visual quiz game. (that I built over the weekend) The application is powered by @commonstack_ai a unified API gateway for large language models. It enables accessing multiple models for planning, image generation, and curation through a single API. One Platform, One Bill. How the game works: • Choose Solo or Multiplayer mode Solo Mode Enter your username, Commonstack API key (Persists Locally), quiz topic, difficulty level, and number of rounds. The system then generates the quiz automatically. Multiplayer Mode The host configures the session by entering a username, Commonstack API key (Persists Locally), quiz topic, optional room name, difficulty, maximum players, number of rounds, and time limit per round. A room is generated, and other players can join using the room code shared by the host. Quiz generation runs in parallel, allowing multiple rounds to be prepared efficiently without significant increases in setup time. You Get 3 Hints. Faster the answer, more are the points. The project is open source. You can host it locally or try it online. Give it a try, links below.

Today, we share a breakthrough on the planar unit distance problem, a famous open question first posed by Paul Erdős in 1946. For nearly 80 years, mathematicians believed the best possible solutions looked roughly like square grids. An OpenAI model has now disproved that belief, discovering an entirely new family of constructions that performs better. This marks the first time AI has autonomously solved a prominent open problem central to a field of mathematics.

How do you evaluate an LLM router fairly? Most benchmarks look at prompts, but routers operate at an agentic-step level. A router that saves money but breaks the task could be worse than no router. We open-sourced TwinRouterBench to measure this honestly. 🧵

@gradientintern Hoppers

H200 and B200 GPUs continue to rise in May as agentic usage takes off. Price two weeks ago vs now: H200: $3.85 vs $7.01 B200: $5.17 vs $5.73 It’s more important than ever that you not only have compute, but can effectively optimize usage so no compute is wasted in this shortage

Are you kidding me??? It grasps multiple objects with different ways, all at once with… a single hand??? No pauses. 1x speed. GENE-26.5 is @gs_ai_’s robotics-native multimodal foundation model. It’s trained on 200,000+ hours of real human hand data (motion, force, touch) and runs on a 54-DoF bimanual system: Scaling that human data 4x lifted real-robot success rates from 16.6% to 65.6% on long-horizon dexterous tasks!! Same model weights, zero fine-tuning for this exact sequence. You know those tiny coordinated movements you do without thinking…? Robots couldn’t reliably do that before. Now they can. Today. This is the video you’ll send to my friends, outside of our bubble, when they say “robots are still just demos.” Congrats to the entire team around @zhou_xian_! Credit: Seen at Zu Wang (@zuwang95) Genesis official announcement for the full story + longer demo: genesis.ai/blog/gene-26-5…) ——— Weekly robotics and AI insights. Subscribe free: 22astronauts.com

This puts heinous ideas into my brain.

Run Claude Code with Commonstack in 4 steps: - generate an API key - set 4 environment variables - run claude - /status to verify Set it up now in 5 minutes with @alex_mirran.

@tryParallax Sure does

mine looks better

We're open-sourcing Asimov v1, a humanoid robot. With Asimov v1, you can build, train on, and make it your own humanoid robot. It's the first step of building a humanoid labor force for the rest of us. Asimov v1 is 1.2 m tall, 35 kg, with 25 actuated degrees of freedom. Structural parts machined in 7075 aluminium and 3D-printed in MJF PA12 nylon. We're releasing the mechanical design and simulation files. Ready for locomotion policy training out of the box. The BOM is open too. Source everything yourself, or order the DIY Kit. All components, ready to assemble. $499 deposit, $15,000 target price. Ships end of summer 2026. GitHub: github.com/asimovinc/asim… Manual: manual.asimov.inc DIY Kit: asimov.inc/diy-kit Most humanoid robots are controlled by the companies that build them. Asimov v1 is built for the rest of us. Build it, test it, and share your feedback with the community.

Heard some people like wheels？😁 Humanoid robots are the ideal form of general-purpose robots (perfect for general AI and human-derived data). They can work without wheels — but they can also have wheels if they want. Whatever works.

Awesome to see @tryParallax’s distributed framework for heterogeneous machines being implemented and serving up inferences! Build and customize your own clusters for AI like never before 🤖 ./ LFG @Gradient_HQ

@gradientintern @tryParallax @Gradient_HQ Wow damn.. I still hold a bag of theta form 2022.. parallax is a banger

To make this work, we adapted Parallax, @Gradient_HQ's distributed inference framework, to run across EdgeCloud's global node network. One API endpoint, model split across many machines, no centralized cluster required.

Qwen3 32B by Alibaba is now live on Theta EdgeCloud as a decentralized on-demand inference API, a large-scale LLM served across community GPU nodes using pipeline parallelism over the internet. 🧵

Kimi K2.6 is a monster, not only matches in performance against closed lab frontiers it’s also very capable being able to do: “architecture scales horizontally to 300 sub agents executing across 4,000 coordinated steps simultaneously” a 3x in sub agents from K2.5 of 100 and 2.5x in coordinated steps of 1,500. Not to mention its long horizon capabilities are incredible with “4,000+ tool calls, over 12 hours of continuous execution, and 14 iterations”. You get all of this and many more features + it’s open source and at a price of 3-5x below its closed model competitors such as GPT 5.4 (xhigh), Claude Opus 4.6 (max) and Gemini 3.1 Pro (thinking high) Absolutely beautiful work

@kimmonismus Aww

That made me laugh and feel sad at the same time.

smh, is there a better way to do this on @MATLAB ... im only halfway done....

Our cofounder @0xEricYang sat down with @yacinelearning to walk through Echo-2’s distributed RL architecture. Dive in to learn about async RL with distributed infra, and how we are scaling this for businesses to win in the agentic era.

When you scale parallel agents, prompt updates degrade fast. The more trajectories you process concurrently, the more generic your learned prompts become. Our researchers worked with @lihanc02 and team on Combee to rethink how aggregation works at scale. Results held up across GEPA and ACE even past 80 concurrent agents. Read more on this research👇