Viktor.comAn AI coworker that actually does the work
Dropstone is persistent AI built to become the intelligence that runs with you. It remembers experience, learns from mistakes, builds skills, verifies its own work, and applies what it learns across code, chat, docs, spreadsheets, and environments. It isn't locked to one model or interface. Dropstone is a runtime for intelligence, one system that can evolve from an AI assistant into an always-on digital workforce.
This is the 6th launch from Dropstone. View more
Dropstone
Launching today
Dropstone is a runtime for intelligence, one system that can evolve from an AI assistant into an always-on digital workforce.
Free
Launch Team
Dropstone
I would like to answer the question before you ask:
1. What does Dropstone do differently from others?
Dropstone is about ensuring that what was learned, corrected, or established in one interaction is carried forward to the next. We do not need to add a human to remember every correction. If a system is corrected or learns something important, we can capture it automatically and use it for future interactions.
We’ve measured 0 → 100% held-out transfer in our internal benchmark and we continue developing and validating these results, and we hope to publish more about our methodology, findings, and what comes next at blankline.org/research. Subscribe to our newsletter there to stay up to date.
Dropstone
2. Is memory alone a feature, not a product?
Memory alone is not enough for a product; we want AI to learn from experience, verify its learning, and apply it consistently in future interactions and across different surfaces. We are building the system around the learning loop: memory, skills, correction, verification, and persistence.
We want to prove that these AI systems are more reliable, more capable over time, and more useful in the real world. We will share more of our research, measurements, and product development publicly on blankline.org/research.
Dropstone
3. Why should I choose you over others?
You don't have to — and that's exactly the point.
We are not asking anyone to choose Dropstone because of a pitch, a comparison chart, or a promise. We are asking you to try it.
You can use Dropstone for free at chat.dropstone.io, or download the CLI at dropstone.io/downloads and create a free account.
Our Fast tier is designed to give you a lot of headroom before you need to subscribe. The best way to judge that is to use it on your own work and see how it performs for you.
We also periodically increase usage limits as our user base and infrastructure develop. These increases are occasional and not guaranteed — they depend on usage, capacity, growth, and what we are able to support at the time.
For example, when we announce a 4× increase, it means that someone who is paying for a $15/month plan can temporarily receive approximately $60 worth of usage, while a $75/month plan can receive approximately $300 worth of usage, and a $150/month plan can receive approximately $600 worth of usage for a period of time.
So the customer is still paying $15, $75, or $150 respectively — the additional value comes from us increasing the usage allowance on that existing plan.
These kinds of increases are something we may offer from time to time as we grow. We don't want people to assume they are permanent or guaranteed; we'll announce them when we make them available.
For updates like these, follow @Blanklineorg and @dropstoneio on X. We may announce changes on either account:
https://x.com/Blanklineorg
https://x.com/dropstoneio
And we've shared examples of these updates publicly as well:
https://x.com/Blanklineorg/status/2063966499837280510
https://x.com/Blanklineorg/status/2072668409293423037
Ultimately, the answer to “why should I choose you?” is simple:
Don't take our word for it. Try Dropstone and decide for yourself.
Dropstone
4. What is the point of using open models with Dropstone instead of using the model providers directly?
The point isn't limited to open models. Dropstone is designed to work with both open and closed models. But we deliberately use open-source models in our own system because we want to be transparent about what is actually running underneath the product.
We don't believe the model alone should be the entire product. A major part of our performance comes from what happens around the model — especially memory, learned skills, self-verification, and the ability to carry improvements across interactions and surfaces.
That means you can take an open-source model and use it through Dropstone, and potentially see a meaningful performance improvement from the system around it. We also don't want users to have to constantly switch products every time a new model is released. Our job is to evaluate the models for you.
We'll continuously audit and evaluate the available models against the workloads we care about. When a newer model performs meaningfully better than the model we're currently using, we'll replace or upgrade it where appropriate.
Here are the Technical Report and Blog For Dropstone 1.7 Models:
Technical Report: https://blankline.org/research/dropstone-1-7-technical-report
Blog: https://www.dropstone.io/blog/introducing-dropstone-1-7
So our philosophy is simple:
We don't need to win the model race. We need to make the models work better for you.
Being model-independent is deliberate. Your accumulated memory and skills should remain useful as the underlying models improve, rather than being locked to a particular provider or generation.
That means you can use open models with Dropstone, use closed models with Dropstone, and let the underlying models improve over time without having to rebuild your entire workflow around whichever model is currently the newest.
We're building for the long term: the models will keep changing; what the system learns from you should keep compounding.
Dropstone
5. If I want to use my memory in my own custom applications, can I?
Yes. Your Dropstone memory and accumulated knowledge can be used across the applications and surfaces you connect to Dropstone.
We built the Dropstone SDK so you can bring that memory and knowledge into your own applications instead of keeping it isolated inside the Dropstone interface.
You can get started with the latest npm package:
@blankline/dropstone-sdk
https://www.npmjs.com/package/@blankline/dropstone-sdk
You'll need an API key, which you can generate from your Dropstone dashboard:
https://dropstone.io/dashboard/settings
For the full SDK documentation and examples, see:
https://docs.dropstone.io/cli/sdk
We also introduced the SDK in more detail in our engineering blog:
https://www.dropstone.io/blog/dropstone-sdk-1-0
The idea is simple: the memory you build with Dropstone shouldn't be trapped inside one interface. You should be able to use that accumulated context and knowledge wherever you build.
Dropstone
Here are a few key features worth highlighting in this thread:
1. What’s in it for enterprise?
Dropstone is built for companies that need the power of AI while keeping control of their data.
Dropstone can be deployed within company-controlled environments, including on-premises infrastructure, private clouds, and highly restricted environments. This means sensitive data can remain within the organization’s controlled environment.
The bigger opportunity is that companies can build and deploy AI systems across their organizations without necessarily sending sensitive underlying data to a public AI service.
For enterprises, the question isn’t just what can AI do? It’s also where does your data go, and who controls it?
For Sales: dropstone.io/contact
Dropstone
2. Can Dropstone control my smart home, make phone calls, or monitor my email?
Yes.
Dropstone is designed to connect with the tools and devices you use—but it doesn't get access by default. You choose what to connect and authorize.
For now, Dropstone can connect with Samsung smart devices, and we're expanding support to more devices.
It can also make phone calls and have conversations with you. Here's an example of how that works:
https://x.com/Blanklineorg/status/2066980979928178977
We also tested this capability through our research project and put it through adversarial testing to see how it performs under challenging conditions. Take a look at Project Yaazh:
https://blankline.org/research/project-yaazh
And yes, Dropstone can monitor your emails. Google integration is currently available to beta users.
The goal is simple: an AI system that can actually interact with your digital and physical world while you remain in control of what it can access.
Dropstone
3. How long can Dropstone think?
Unlike systems that are typically optimized for short response cycles, Dropstone can sustain reasoning for minutes to hours, depending on the complexity of the problem and the reasoning effort selected.
In our testing, when we set the reasoning effort to MAX, Dropstone sustained its reasoning process for 2 hours on a single problem.
And we found something interesting during that test.
We're currently preparing the results as a research publication, which we'll be releasing in the next few days at: blankline.org/research
More to come.
Go Mail Merge
@santosharron awesome
Dropstone
How can I join the community?
If you want to be part of what we're building, you can join us on Discord:
https://discord.gg/RSrPutXpXd
That's where we share updates, discuss what we're building, and hear directly from people using Dropstone.
You can also sign up to receive updates by email and stay informed as we release new research, products, and announcements.
We're building Dropstone in public, and the people using it, testing it, sharing feedback, and pushing us to improve are an important part of that process.
Product Hunt Wrapped 2025
Congrats on the launch day!