Big milestone for Reducto: we’ve raised a $75M Series B led by a16z, bringing our total funding to $108M. Since our launch, we’ve been obsessed with pushing the limits of what’s possible when AI meets the real-world messiness of documents. We’ve now processed 1B+ pages, nearly 6x the pages since our Series A less than 6 months ago. We’re doubling down on model research, developer tooling, and the reliability that production AI demands. We'll be announcing a series of product improvements that will extend our capabilities even further, helping our customers get even closer to perfect accuracy across their most challenging docs. A big thank you to everyone who helped get us here. But we're just getting started - if you're interested in what we're doing, we’re hiring across all functions! reducto.ai/careers

Jev doesn't support image input yet, so I tested out some open-source Jev-like alternatives. Seems like DiffusionGemma and reflex are on the pareto for the open source Vision-Jevs! github.com/vllm-project/vllm… (ty @mmastrac) github.com/kshetrajna12/refl… (ty @kshetrajna )

with instinct, you can now sign up for @reductoai and use r-1 to find out how bad your finances are in 20 seconds

your agents can sign up and use reducto totally headless! give it a shot today :)

I think you can get pretty close with a single qwen 0.8B forward pass with constrained decoding to a single output token. the "output tokens are free" thing you get kinda for free with prefix caching (you only do the prefill once for each query). it's still a pretty neat idea and they obviously executed it well

a bunch of tools require expensive enterprise upgrades to support SCIM, but most of them have APIs for user management. so you can just have your agent deploy a cloudflare worker that runs a poor man's SCIM sync for you keeping the user/groups in line. no upgrade required!

I have the same qualm - I feel like for the "trusted inner circle" people I just want it to coordinate without needing permission every time

I was encouraged this week to see the leaders of the frontier labs agree on the need for them to slow down the pace of AI development. Given the stakes, it’s a good and necessary first step.   But I’m even more encouraged by the growing recognition that how this powerful new technology develops should be at the center of our public debate.   I’ve been watching the progress on AI for over a decade now, and one thing that’s clear to me is that the potential impact of this technology is not overhyped. It’s also moving at lightning speed – and even faster than those who are engineering it can keep up with.   I’m not an AI accelerationist who believes it will lead to some techno-utopia, and I’m not a doomer who thinks it will inevitably lead to humanity’s destruction. But whether this technology results in amazing breakthroughs in medicine, energy and education or unleashes huge economic disruptions, greater inequality, and potential catastrophe will depend on the choices that we make right now – choices that should be made not just by the companies involved, but by all of us.

first thoughts on @Muse - they went the enterprise route and ask for permission way too much. asked it to cancel a day of meetings and I need to click approve manually for each one. what even is the point of using AI at that point. no bypass permissions mode available.

i’m continuously shocked by how hard it is for enterprises to get GPUs in 2026 shocking number of them just flat out refuse to provision any regardless of the cost or efficiency savings they’ll see as a result

have had to start using /goal more with astra - it just stops randomly all the time for no reason

what is it with companies building an in product agent with access to more tools than their public api or mcp. especially for tools geared towards technical users, please just let me use my own agent.

great UX @incident_io! was delightful that this worked out of the box and the dynamic UI screenshot after was excellent

look at your data is important. ask your coding agents to look at the data != look at the data yourself

We're making Extract even better! It's now twice as fast, more accurate, and a fraction of the cost. What's new: → Faster v4 model with better schema adherence & up to 99% citation coverage → Flat 2¢/page pricing, all-in. No added parse costs. → Upcoming MCP tools to help teams set up and optimize extraction schemas The team @reductoai has a lot more coming soon — stay tuned!

he can't keep getting away with this

I promise that we will do everything to: - Make Reducto the best product in the industry - Get it in the hands of all of the companies that need it - Do anything and everything to make those companies successful Momentum is increasing but we're at a fraction of our ambition