UltiHash

Store data for AI models, using less space, with high speed

119 followers

Store data for AI models, using less space, with high speed

119 followers

Built to meet the demands of AI and advanced analytics, UltiHash reduces storage resource use by up to 60% while delivering the throughput needed for large-scale data operations. With byte-level deduplication, fast deletion, and seamless integration with cloud and on-prem infrastructures, UltiHash provides a scalable storage solution for data-driven enterprises.

Free Options

Launch tags:Artificial Intelligence•Data & Analytics•Database

Launch Team / Built With

CloudflareYour next app belongs on Cloudflare Workers. Deploy for free

Promoted

Maker

📌

Excited to introduce UltiHash, the next-gen object storage platform! 🚀 Built for AI and analytics, it cuts storage volumes by 60% with byte-level deduplication. UltiHash ensures data resiliency with erasure coding and includes advanced access management for security. You can deploy seamlessly via Kubernetes with S3-compatible integration, whether in the cloud or on-prem. Plus, our efficient data deletion process frees up capacity without slowing down the system. We can’t wait to hear your feedback and see how UltiHash supercharges your projects!

Report

2yr ago

Hi Tom and UltiHash UltiHash looks cool – I’ve installed the minikube setup and already got started. Could you please explain in more detail how the space savings work? I am working with vast amounts of data, primarily video data, compression needs def upcoming

Report

2yr ago

Maker

@jan_renken Hi Jan, Good to hear that you set up Ultihash :) Our deduplication process works by splitting objects into bytes fragments of varying sizes. If a fragment already exists within the system, it isn’t stored again, eliminating unnecessary duplication across datasets. This ongoing comparison ensures that storage resources are utilized efficiently while maintaining data integrity. You can find out more in our documentation: https://docs.ultihash.io/1.-prod...

Report

2yr ago

Cool product. AI world is slow and expensive, amounts of data grows rapidly, new huge data domains are here (video, multimodal) acceleration is needed at all layers. Glad to see products like Ultihash!

Report

2yr ago

Maker

@denis_timonin thanks for the feedback!

Report

2yr ago

Really great to see innovation happening at the storage level. It’s often overlooked as “good enough”/“solved” as part of the standard cloud offerings, but could hide significant improvements, esp. for new-ish AI/ML use-cases. Nice job, UltiHash team!

Report

2yr ago

Maker

@alex_bbb thanks Alex, much appreciated to read your thoughts on UltiHash!

Report

2yr ago

Maker

Super proud to be part of the team that built UltiHash! Enabling companies that want to innovate with AI to do it without suffering from storage bottlenecks by saving resources is really cool - especially when we think that data storage produces more CO2 emission that the aviation industry...! Can't wait to get more users' perspective on UltiHash :) On a personal note: amazing to see a project that goes from an idea to an actual product. It's been quite the journey, getting into the startup and the tech worlds - 100% worth it.

Report

2yr ago

Maker

@juliette_lehmann great working together Juliette!

Report

2yr ago

Very much needed! We’re talking to so many AI leads at corporates & industrial companies and they share that what they really need are products that enable them to build better, faster and cheaper AI models instead of buying off-the shelf AI solutions on the application layer. We need more boldness like this!

Report

2yr ago

Maker

@linh_s Thanks so much! That’s exactly the gap we’re aiming to fill—helping companies build AI models better, faster, and with fewer resources. It’s great to hear that this resonates with what you’re seeing in the market. Here’s to empowering AI development!

Report

2yr ago

Langfuse

congrats on the launch! love the deduplication as a lot of the blobs our users store includes large duped blobs of context shared across llm invocations. s3-api is awesome to see as well.

Report

2yr ago

Maker

@marc_klingen Thanks Marc! Great to see you’ve noticed that bottleneck with LLMs. That’s exactly where UltiHash steps in—enhancing storage efficiency without slowing down operations.

Report

2yr ago

1 2

Reviews