
Q3 was our best quarter ever. Woo! [Please forgive the lack of a Y-axis]

Co-Founder & CEO of @JoinVantage: infrastructure cost management.

Q3 was our best quarter ever. Woo! [Please forgive the lack of a Y-axis]

Exciting launch: Today, we're introducing Vantage Tokenomics. Every company I talk to right now has the same problem. AI coding spend is climbing fast, and they can see the bill, but not what it bought. One engineer might be spending 10x the rest of the team. Are they shipping 10x as much, or using tokens 10x less efficiently? Until now there wasn't a good way to answer that. Vantage Tokenomics attributes token spend to three things: identity, purpose, and outcome. You're given a full company roster showing everyone generating token spend, ranked by estimated spend alongside key metrics like @GitHub pull requests or @Linear issues they've completed. Click someone's name and you see every task they've finished and what each one cost. Vantage Tokenomics will also automatically classify each request with tags as it happens with who made it, what it was for, and the PR, branch, repo, or ticket it maps to. You can also customize this based on your organization's own tagging strategy. Nobody has to tag by hand. As part of this suite, we're also offering Tokenomics reports which brings in efficiency metrics like cost per million tokens, cache hit rate, and input/output ratio, so teams can find spikes and coach better usage within minutes instead of at month end. It's built on OpenTelemetry, so you can track requests across *all* IDEs and major Gateways out of the box. Cursor, Claude Code, Codex, and the major AI gateways have day 1 support. No applications are required to install on developer laptops, and prompts and completions are never stored - we've taken a thoughtful approach for security considerations as part of this. This is among the biggest launches we've done and I'm really excited about it. If you're interested in getting access, sign up here: vantage.sh/lp/tokenomics?utm…

AI is just cloud all over again. Linked billing on the Claude Platform - excited to see what comes of this :) claude.com/resources/article…

A new level to going full yolo
Selling Oculus to Facebook in 2014 gave me money for helicopter flight lessons and developing a seastead called SLAB with a 1,000 year design life. The goal was to be cheaper than land and absent state/local laws that cripple critical American industries like metal processing, medical research/treatment, desalination, oil refining, chemical engineering, and more. Me and a few close friends cashed it up at a $1B valuation back in 2017, and while it still has a long way to go, there has been significant progress towards my goal of reviving manifest destiny and westward expansion. At a time when most of the seasteading community wanted to escape the US and smoke pot, this was the one project explicitly designed to make America stronger, a relief valve for ill considered or outdated rules and regulations. I am only talking about it now because a few media outlets got wind of me and some friends topping up the coffers and plan on calling SLAB my new floating AI datacenter project. It is way cooler than that, way older than that, and I hate looking like a hype-chasing consensus loser, so now you have to read this tweet.

Launch day: @JoinVantage is launching Inference Allocation! More teams are running open-weight and fine-tuned models on @baseten, @togethercompute, @modal and Bedrock. Those are billed by the hour, not the token. Your provider bills you for the GPU hours whether the model served ten requests or ten million, so the cost of running these models showed up as one shared line item. Now Vantage takes the request logs you already send us (via OpenRouter, LiteLLM, Cloudflare AI Gateway, Bifrost, Bedrock invocation logs, etc) and splits each model's full daily cost, idle time included, across the requests it served. You set weights for input, output, cache read and cache write tokens based on the compute they take, and allocations always sum back to what the provider billed. We put together a quick video on how this works below. vantage.sh/blog/inference-al…

Everyone’s out here living the same life.
My coworker spotted me taking a 7pm call like this

Vantage is now an observability provider directly available in the @openrouter console! Very excited by this and have some interesting things coming around it soon.

Launch day: @JoinVantage now supports @CoreWeave costs. For teams that train or serve their own models, GPU capacity is often the biggest part of the AI budget. But CoreWeave usage lived in the CoreWeave Cloud Console. Getting it next to AWS or your model providers meant exporting usage, pricing it against your contract by hand and uploading it as a Custom Provider. Now you connect a CoreWeave organization with an API access token and we import usage every day by SKU, cluster and region. CoreWeave's FOCUS export doesn't include costs yet, so we price it with a per-SKU rate sheet that starts at CoreWeave's on-demand list prices and that you can update to your negotiated rates. Then use Virtual Tags to allocate each cluster's cost to the team that owns it. Available today for all Vantage customers. vantage.sh/blog/coreweave-co…

Launch day: @JoinVantage now supports @togethercompute costs. Teams use Together AI for open-model serving, dedicated GPU endpoints and fine-tuning, often right next to AWS, GCP or Azure. But until now that spend lived in Together's own cost analytics, separate from the rest of the AI budget. Now you connect a Together AI organization with an API key and we pull in daily costs by product, model, GPU, project and endpoint. Dedicated Inference is broken out by GPU type and on-demand vs reserved GPU-hours. Serverless Inference is broken out by model and input, cached input and output tokens. All of it sits next to your other AI providers and the rest of your cloud bill, with Budgets, Anomaly Detection and Virtual Tags. Available today for all Vantage customers. vantage.sh/blog/togetherai-c…

🤔
JUST IN: Sam Altman warns of short-term turbulence saying that some neocloud AI demand was double-counted.

Picked up a raspberry pi after work, worked with Muse to write a python file to persist the data from the Aranet4...it works...and to state the obvious: AI is magical. Setting up a regular cron job to cache the data for the long term now. What's behind this is a long, tiresome, righteous legal battle with my apartment building that you can see I will not lose :)

So happy these guys are making it. Such a crazy scene.
Pierwsze po wybudzeniu słowa bohaterskiego policjanta Artura Morańskiego: „A Kuba żyje?”. Pytał o Jakuba Kulika, drugiego z bohaterów. Takie chwile przywracają wiarę w człowieka.

Got an Aranet4 for my apartment - coolest gadget I’ve bought in a while.

What the helly
7 months ago, AOC and I introduced an AI data center moratorium. Last night, 43 of 44 communities voted for one locally. Americans don’t want Big Tech oligarchs wrecking the environment, gutting jobs and threatening humanity’s future. NATIONAL DATA CENTER MORATORIUM NOW

👀
NEW: Apple is cutting employee AI spending budgets even as it pushes workers to use the technology more. Engineers have seen daily limits cut from $300 to $200, with further reductions to $100 planned. Other teams could see limits fall to $25. bloomberg.com/news/articles/…

We have an ever-increasing number of new Vantage accounts being created by AI agents...so we're making it a bit easier for them to get going. We just added devoted endpoints and documentation specifically for agents to create new Vantage accounts: docs.vantage.sh/auth