Pacific retweeted
We built doc extraction that owns the Pareto curve for almost 2 orders of magnitude
Cost efficient structured outputs:
- Near frontier quality structured extraction
- Up to 99% cheaper
- Up to 3x faster
We've been using it under the hood at Pacific. Today we're releasing it broadly as a research experiment.
Introducing nerfguard-extract-1
extract.nerfguard.com
Pacific retweeted
Wow! This has continued to blow up more than we anticipated.
Y’all melted some servers over here.
We shipped a handful of changes recently to improve stability and make routing behavior more understandable and predictable.
You now can have much more control over model routing — including the ability to get up to 1.5x usage on a chosen model in Claude Code or Codex without the need to route.
Research preview install: nerfguard.com
Highlights below
We built a tool to increase our Claude Code and Codex usage by 3x with the same spend.
nerfguard.com
It also saves us hours a day per person in time we’d otherwise be waiting around for model generation time. Today we're sharing it with the community.
We've been utilizing this internally for a while now and we're excited to share it as a gift with the community. We've seen ~3x more coding agent usage on the same spend and hours saved per person.
We built a tool to increase our Claude Code and Codex usage by 3x with the same spend.
nerfguard.com
It also saves us hours a day per person in time we’d otherwise be waiting around for model generation time. Today we're sharing it with the community.
Pacific retweeted
Such a great team to be working on such an important problem!
As everyone in the space knows, context management is a key missing layer in unlocking the potential of ever-more-powerful models for the use cases that matter most across the workplace and home.
Pacific retweeted
Noah gets a ton of credit for the ad-supported Amp Free idea and helping us get it off the ground. He's an independent thinker and can make crazy stuff happen.
Congrats to him and the team on launching Pacific!
Every company and every person will one day have a magical library that personalizes their computing experience. We're the team building it.
Meet Pacific:
To start, we’re introducing the world’s first enterprise context management system.
To serve the world at large, it’s sometimes best to start on “hard mode.” We've been working with some of the largest companies for which information governance, portability and reliability is of the utmost importance — leading financial services firms — to turn their internal knowledge into a competitive advantage.
To outrun their competitors with AI that cannot only leverage raw intelligence, but the information that makes their company their company. And do so in a monitored, permissioned and secure way that leads the industry on speed and quality.
Just as AI foundation labs put raw intelligence an API call or query away, Pacific does the same for your internal data.









