Can't believe I'm actually profiting off my prompting skills on SWAN 🦢 😂 This AI Agent inspired by the Saratoga guy just bought 2 of my prompts to help its lore develop. Earning crypto doesn't have to be complicated guys. Stop crying over charts and just prompt instead.

We are building a global silicon fabric, integrating CPUs, GPUs, and NPUs to create the biggest decentralized AI Accelerator: * 30,000+ nodes * 15,000+ active nodes per day * 50M+ inference Learn more about Dria Launcher—over half a million downloads—and how to contribute 👇

Join us for a deep dive into open-source AI, RL, commodity hardware inference, and next-gen silicon. May 7th — we’re hosting: - Kyle Corbit (Kyle Corbitt) from OpenPipe, sharing insights on reinforcement learning and ART, an open-source RL library for agentic AI. - Peter

Most of the world's energy flows through heterogeneous hardware, yet today's AI software is designed for homogeneous and centralized setups. The concentration of high-performance AI into the hands of a few powerful entities poses profound risks to humanity. Introducing Dnet: a

AI hardware/infrastructure loves OR;

This will be dnet one year from now.

within a set of 50 problems, it matched the sota 75% of the time and exceeded it 20% of the time

LLM inference is too slow, too expensive, and too hard to scale. 🚨 Introducing llm-d, a Kubernetes-native distributed inference framework, to change that—using vLLM (vLLM), smart scheduling, and disaggregated compute. Here’s how it works—and how you can use it today:

1/ after some rigorous analysis, we learned that some of the “optimal” design choices in dePIN are anything but. in this post, we use optimal control theory to shed some light on dePIN emission policy and beyond

sorry y'all, inference gonna be the dynamic training you're looking for

Launching Decentralized Batch Inference—cutting costs by 5x+ vs. traditional providers—and hitting 20,000 daily active users contributing hardware globally. Dria Batch API is perfect for synthetic data, offline model eval, and cost-sensitive batch jobs.

Introducing our research on low-overhead, privacy-preserving distributed inference: Protecting inputs, outputs, and model weights using lightweight permutations, enabling efficient and privacy-preserving LLM inference at scale.

"Be as casual and as long as you want" **entropy increases**

NYC July 1: Join us to explore private/distributed inference & agent-native platforms. Talks: Brandon Reagen (FHE Framework for Deep Learning), Shubham Toshniwal (Synthetic Data), Aaron Wright (AI Driven Venture Capital), Stefan Teleman (Distributed Inference).

DriaConnect drew more than 200 attendees to Betaworks NYC last week. Explore sessions on agentic AI, synthetic data, distributed inference, and private AI featuring researchers from NVIDIA, NYU, LastMile, and more below.

Introducing the Inference Arena LLM benchmarks are scattered and inconsistent. The Inference Arena ends the guesswork by letting you explore, compare, and customize benchmarks that match your exact workload. No spreadsheets, no hunches, only data driven choices.

Introducing our latest research on open evolutionary agents: We conducted 29 large-scale experiments using OpenEvolve and AlgoTune to test how evolutionary agents improve algorithm performance through iteration, tuning, and model selection. Here's what we found:

DIY AI: Own Your Intelligence, Istanbul. Builders in İstanbul, join us at Arkhaus on Sept 2 as a part of Builders Week Istanbul (Sept. 1-7) We'll discuss: - Verifiable agents w/Mertcan (EigenCloud ) - Decentralized inference and evolutionary AI w/andthattoo (Dria) - The

LLMs are stateless. We built Dria Mem Agent to change that: Making memory a first-class feature. A 4B agent with local interoperable memory across Claude, ChatGPT and LM Studio. It turns LLMs from stateless chat into stateful agents with persistent human-readable memory.

Tonight in SF, we’re under 6 hours from kickoff and sitting at 350+ registered. If you want to deploy multi-agent systems with high accuracy and low cost this is the event to attend. Particularly relevant for those working on DSPy, real-world evaluations, and compute.