a few notes on local AI and our computers that run it around fifteen months ago, i led the pre-seed round for a weird local AI project it was called @actualinc led by @Tom_A_Lynch for the first time in the AI movement open models are now good enough for a large and growing set of workflows, especially where latency, privacy, bandwidth, and token cost are critical hermes agent from @nousresearch and ‘lab’ by @primeintellect have broken this idea wide open with killer products today, built on open rails we are no longer forced to assume that every useful task must route through one of a few centralized frontier APIs and so tom and the squad at @actualinc have painstakingly created a new type of proprietary inference engine for all types of computers maybe you are an owner of one of the tens of millions Apple macs or an owner of one of the millions of RTX-class machines or maybe you run a small team or company and you are starting to realize what many others across the world, have started to realize which is that sometimes centralized AI systems bring their own set of unique problems especially as our workflows move into an always-on, agentic era which might sometimes come with: - exorbitant costs - no privacy - lack of control but a problem remains running local AI is still too hard firing up one model on one machine is difficult enough but running models across different machines means navigating drivers, networking, model sizing, memory limits, API setups stitching together hardware is messy across Mac, Windows, Linux, NVIDIA, Apple, AMD heterogenous systems were never meant to work seamlessly together like this today, you'll find that most of the local inference software available is built around single users, for a single machine use case running inference capacity in bulk, across many users, and many kinds of computers that is a different animal altogether tools like llama.cpp made local models accessible, but they are not designed to coordinate tens of thousands of heterogeneous machines under a live supply network vLLM-style batching can be economical, but vLLM is limited in the machines it can run on the local AI movement requires a simple product experience, for a tough engineering problem that is the problem @Tom_A_Lynch pitched me fifteen months ago and the problem @actualinc has been engineered to solve now with the help of his cracked and growing team @skooookum, @somewheresy, @jack15032, and the rest of the crew starting with their first product go to actual.computer in under one minute, install the @actualinc client, authorize your machine, load a model, and @actualinc turns your personal device into a private inference endpoint in less than twenty seconds, add a second, or third, or fourth machine of any type and it will start behaving like a small private cluster from a technical POV actual runs as a lightweight daemon on each machine and turns local hardware into a routable inference endpoint each node reports its available memory, accelerator type, model state, and health back into the cluster, while the user interacts with a single API surface instead of managing ports, drivers, or SSH sessions a relay layer lets requests reach machines without forcing users to configure networking or expose their devices directly and as a result computer owners of all kinds of computers can get local inference clusters - when they want to control costs - when they want to control privacy - when they want to build on top of a neutral stack eventually, over time computer owners running these local clusters should be able to seamlessly contribute their unused capacity back to the @actualinc network for a market to price and for someone else to purchase you can think of @actualinc as building the engine and control plane for datacenter style throughput in homes, in offices, in small to medium sized businesses, and in very big sized businesses because there are hundreds of gigawatts already on the grid right now, that are waiting on desks, in storage closets to become actual computers

thank you g, appreciate you a few random notes about this topic from my time running @collab_currency in general i remain a big believer that we should be using the internet and blockchains to open up better access to ideas, projects, and opportunities at the very earliest stages stuff like fair launches, PoW, and cap formation tools all of this extends from the same design principle of more equitable ownership formation which i am very much inspired by and oriented toward all that said teams at the earliest stages still need a combo of tools in the toolkit for a higher probability of success the three i usually point to are: - capital - connectivity - counsel capital gives founders room to hire, experiment, build connectivity is the invisible and visible networks that help founders better find users, first hires, customers, collaborators counsel is the experience from others around the project that can help founders work through hard decisions with people who’ve seen similar challenges before so all this to say at the very earliest stages, crypto/blockchains have started to become an amazing tool at pooling capital together for founders and projects but i have found that it has had a harder time with the other two (connectivity, counsel) which is the stuff after the capital arrives i am talking my book a bit but i'm also pretty optimistic that these two don't necessarily have to come from a pre-seed/seed stage investor in the future over time i’d love to see those parts of the venture model unbundled until then, i think i think there's a tremendous amount of value to be created showing up for founders at the earliest stages helping them navigate their project to higher ground standing behind them as they do the tough stuff

v excited for @roborobots to come out of stealth a new robotics company based in los angeles have known @gpudad and @kylenoble a long time they are high integrity, big hearted, brilliant dudes who just seem to be wired up to solve difficult engineering problems historically across stuff like blockchains, software, hardware, AI robo is built on v simple idea beautiful, low cost robots, owned end to end, deployed fast in the real world, for a wide range of outcomes low cost robots mean accessible deployments, more deployments means more data, more data means model improvements, better models improve robot costs/performance, better robots unlock more novel deployments robo's robots are rolling out across apparel warehouses, hospital pharmacies, supply chain logistics, food manufacturing i hope to see them all over LA soon gg

killer work from the @primeintellect crew now crossing $100M ARR a massive vibe shift for anyone paying attention it is v clear at this point that the AI landscape is undergoing a radical transformation we are moving away from the world of 'renting intelligence' from the largest labs to a new world where everyone must own the workflows, data, and optimization loops that will make your unique intelligence compound prime intellect is building the open infra stack for all types of companies to train, evaluate, deploy, and continuously improve models they own from compute and RL to environments, sandboxes, evals, and deployment all over credibly neutral architecture deeply proud of this crew, a true david vs goliath story to building across this new AI landscape our squad at @collab_currency are proud investors in every round since the v beginning of the journey massive kudos to @vincentweisser, @johannes_hage, @jimmy_811, @willcb and the rest of the prime squad the future of AI will not just be built inside closed labs it will be built by teams with the tools to own their own intelligence gg