Sanctum Profile Banner
Sanctum Profile
Sanctum

@SanctumAI

3,932
Followers
12
Following
41
Media
154
Statuses

Run private AI on your desktop 🖥️  Fully encrypted, stored locally ✨ Join our community:

Download app 👉
Joined May 2018
Don't wanna be here? Send us removal request.
@SanctumAI
Sanctum
1 year
Your conversations with ChatGPT are not private 🔓❌ That's why we built Sanctum, a secure and fully encrypted desktop application that lets you run LLMs locally. Today we are excited to roll out the first version of Sanctum into the wild. Let's unpack what Sanctum is 🧵
Tweet media one
7
22
78
@SanctumAI
Sanctum
10 months
🚨 New release 🚨 We are only a few steps away from introducing a paid version of Sanctum. And we are excited to announce early access to some Sanctum Pro features for free for the next few weeks. Run Mistral LLM locally, experience up to 5x faster AI responses, and more 🧵👇
2
10
38
@SanctumAI
Sanctum
9 months
Introducing Sanctum Pro – the easiest way to interact with open-source LLMs 👑 🗳️ Secure, 100% Private PDF Chatting 🔍 Advanced Sanctum Vault Search ⏳ Early Access to New Features ... and more on the horizon
6
6
30
@SanctumAI
Sanctum
1 year
Hello world ☀️
2
6
23
@SanctumAI
Sanctum
5 months
We've just launched the quantized 8B version of @AIatMeta LLaMa 3 on @huggingface Pick the version you want to run here 👇
Tweet media one
1
11
26
@SanctumAI
Sanctum
1 year
Sanctum is available today for Macs. The app works best on M1 (or later) Macs with 16+ GB memory. Windows & Linux support coming soon! Experience the power of generative AI running directly on your device. Set up your Sanctum now 👉 download the app from the link in our bio.
1
0
25
@SanctumAI
Sanctum
1 year
Sanctum believes the future of AI is local. Running apps like digital assistants shouldn't mean sacrificing privacy. With Sanctum you can: 📶 Chat with AI without the internet 🚀 Launch full-featured open-source LLMs in seconds 🔐 Maintain control of all your data
1
1
24
@SanctumAI
Sanctum
8 months
We've got some big news 👀 Adding Windows support in the next release. If you'd like to test it early - join our Discord and become a beta-tester 👇
Tweet media one
1
3
20
@SanctumAI
Sanctum
7 months
Sanctum is now available on Windows devices for free! 🟣 Check out your new local encrypted chat assistant with file upload. Link below👇
6
8
20
@SanctumAI
Sanctum
7 months
💫 Exciting news! Sanctum v1.3 now support Windows ⚡ What's New? • Windows Ready: Download and run Sanctum on Windows (with CUDA 12 support). • PDF Chat Viewer: Seamlessly view PDFs in the Sanctum UI and see how your queries are answered from highlighted text. • Expanded
0
9
22
@SanctumAI
Sanctum
11 months
Ask me something ✨
Tweet media one
3
3
22
@SanctumAI
Sanctum
4 months
Excited to be part of the @huggingface local apps family! 🤝
@julien_c
Julien Chaumond
4 months
A second batch of local Generative AI apps are now live on compatible model pages on @huggingface 🔥 Welcome to the family, friends 🥰
Tweet media one
5
24
148
0
6
21
@SanctumAI
Sanctum
9 months
✨ Introducing Sanctum v1.2.1 🌟 Drag & drop PDF files with ease 📊 Monitor your device performance 🚀 Explore new model: Microsoft's Phi-2 View full changelog 👇
Tweet media one
0
6
18
@SanctumAI
Sanctum
8 months
Thrilled to share the @rootscan_io with everyone ❤️
@rootscan_io
Rootscan
8 months
1: We are excited to announce @rootscan_io as the premiere block explorer for @TheRootNetwork . “We” (here) is the @SanctumAI Labs team where we plan to build additional infrastructure for @futureverse ecosystem. @bornreadyvc invested in us, & we are dedicated to support Root.
20
50
211
0
4
18
@SanctumAI
Sanctum
10 months
Chat directly with us in Discord to be involved with the early product features and more. Join Discord server👇
Tweet media one
1
4
16
@SanctumAI
Sanctum
1 year
Privacy is your fundamental right. Sanctum gives you the power of generative AI right on your desktop, without compromising on your data security. 🔒
0
3
18
@SanctumAI
Sanctum
5 months
Sanctum v1.4 Update 💥 We’re excited to announce Sanctum v1.4 release, here's what's new: 🔗 HuggingFace Integration Download any open-source .gguf LLMs on HuggingFace directly from the Sanctum app for your specific use case. 🦙 LLaMa 3 Support Download and run the latest
Tweet media one
3
7
18
@SanctumAI
Sanctum
7 months
We're very excited to open up Sanctum as a 100% free application for the foreseeable future 🎉 We have a number of exciting updates on the horizon, including a big release today, & want to give as many people as possible the opportunity to use the app & create their own private
1
5
17
@SanctumAI
Sanctum
10 months
Mixtral-8x7B available & running on Sanctum ✨ Download & run just in a few clicks. ✔️ MacBook: M1 Pro Max ✔️ Memory: 64GB (32GB+ recommended) Live demo 🎥👇
0
2
17
@SanctumAI
Sanctum
3 months
We are extremely excited to be listed as one of @NVIDIAAIDev 's partners building with their RTX AI Toolkit! An incredibly helpful platform supporting us as we build out our local fine tuning capabilities for AI. More Information in the link below:👇
0
4
17
@SanctumAI
Sanctum
10 months
Check out Sanctum in the latest @mreflow 's FutureTools newsletter edition "How to build apps without coding"👇 We get a ton of value out of this newsletter. It's a place where we find a lot of AI tools to research and we're honored to be featured ✨
Tweet media one
0
1
15
@SanctumAI
Sanctum
1 year
Setting up Sanctum is simple and secure. Create a local user account and password to encrypt all your data within the Sanctum Vault - your personal storage for conversations. Your Sanctum Vault always stays local. Your are in complete control of your data.
1
0
17
@SanctumAI
Sanctum
1 year
To begin, Sanctum works with one of the most popular open-source LLMs: 🦙 Llama-2 (7B parameters) Support for models in GGUF format coming soon! This will make it super easy for us to support more models down the line.
1
0
17
@SanctumAI
Sanctum
2 months
Gemma 2 2B IT quants uploaded to HuggingFace ✔️
Tweet media one
0
4
18
@SanctumAI
Sanctum
1 year
We'd love to see this too & hope it extends to AI. That's why we are on a mission to bring users the power of generative AI while ensuring their privacy 🫡
@VitalikButerin
vitalik.eth
1 year
@nimivashi15 One of like 10+ ideas, but: I would love to see coordinated adoption of free / open / decentralized / privacy-friendly tech go much deeper than just zupass and zupoll. Use ETH payments on L2, dump telegram, use @ethstatus or @skiffprivacy or other alternatives.
29
58
247
0
2
16
@SanctumAI
Sanctum
4 months
We have successfully fine-tuned a Llama-3-8B-Instruct model on a custom dataset for function calling to the CoinGecko API. One of our lead developers, Andrii Kotliar, stated “Creating the dataset took some time as we figured out how to ensure the model responds appropriately
@NVIDIAAIDev
NVIDIA AI Developer
4 months
🎉 Introducing our NVIDIA RTX AI Toolkit 👉 A new collection of tools and SDKs for Windows application developers to customize, optimize, and deploy AI models to 100M+ RTX AI PCs #COMPUTEX2024
Tweet media one
2
17
87
1
5
14
@SanctumAI
Sanctum
7 months
. @GoogleDeepMind Gemma 7B is now available in Sanctum✨ To run it locally - upgrade to the latest v1.3.1 version of the app.
Tweet media one
0
2
13
@SanctumAI
Sanctum
4 months
Sanctum v1.5. Release 💥 We are excited to share a new release update with new features, improvements, and bug fixes! Here’s what’s new: 📂 File Manager (Beta) This is the first step to our comprehensive RAG solution. View all imported files in a single place, use it to
0
4
12
@SanctumAI
Sanctum
7 months
Update: We are at 17 upvotes on @ProductHunt ! Can we get 7 more upvotes?
Tweet media one
@SanctumAI
Sanctum
7 months
We are currently at 12 upvotes on @ProductHunt ! 😼 Lets double it! Can we get to 24? Link below👇
Tweet media one
1
2
7
0
2
11
@SanctumAI
Sanctum
5 months
Sanctum v1.4.4 is live 💥 - Updated llama.cpp to support pre-tokenizers in new models. - Resolved a bug that caused system prompt words (e.g. <context>) to occasionally appear in AI responses.
1
3
11
@SanctumAI
Sanctum
2 months
Today we're excited to release Sanctum v1.6 with a new UI and a set of new features for the dev community ✨ 🎭 New UI With simplified navigation and updated interface Sanctum gets better, making room for even more exciting features. 🗄️ Local API Server Powered by llama.cpp,
Tweet media one
0
6
14
@SanctumAI
Sanctum
11 months
Which AI models would you like to see added to Sanctum? 🤖
Mistral
11
Vicuna
1
TinyLlama
1
Other (Comment Below)
0
0
5
10
@SanctumAI
Sanctum
1 year
@Web3AlertsTrack
Web3 Alerts | Smart Tracking
1 year
🔔 @DeFi_Dad is now following @SanctumAI Tracking who top Crypto Twitter accounts follow is huge alpha. Track your own list of accounts at
0
0
1
0
1
10
@SanctumAI
Sanctum
9 months
We are now live on Find My AI Tool! 🟣 Thank you for featuring us!
0
1
10
@SanctumAI
Sanctum
5 months
Sanctum v1.4.1 is live ⚡️ In this minor release, we have: - Added preset for Microsoft's Phi-3-Mini model. - Added Account ID to User Settings for enhanced personalized assistance. - Fixed bugs and made improvements.
2
1
10
@SanctumAI
Sanctum
5 months
Thanks @VentureBeat for featuring us & showcasing our truly private, AI for all. @SanctumAI enables businesses & consumers to run Open-Source LLMs locally, securing sensitive data away from cloud vulnerabilities and giving users full control of their data. Read more below:👇
1
5
10
@SanctumAI
Sanctum
4 months
Sanctum v1.5.1 🛠️ We have released an upgrade that resolves an issue that was preventing Sanctum from running on MacOS Intel 🫡
0
0
10
@SanctumAI
Sanctum
8 months
@TechCrunch Sanctum doesn’t share any of your conversations with anyone or the internet. It keeps everything locally on your device so you can use LLMs with personal or sensitive information you don’t want leaked.
2
2
9
@SanctumAI
Sanctum
4 months
Unlock the potential of AI while safeguarding your privacy. Download LLMs like Mistral on Sanctum and join a community that values both creativity and data security. 🚀🌐 Download today! 👇
0
2
9
@SanctumAI
Sanctum
9 months
What are some cool prompts you have used so far in 2024? 🎉🍾 Drop them below and share! 👇
1
1
8
@SanctumAI
Sanctum
5 months
P. S. Shipping direct HuggingFace integration tomorrow so you can run it privately in Sanctum 👀
0
0
8
@SanctumAI
Sanctum
5 months
The evolution of AI is as much about innovation as it is about investment. Check out this snapshot of training costs for major AI models over the years, courtesy of @BayAreaTimes . Our dev @Andrey_best_2 has crunched a training cost estimate for the upcoming @AIatMeta LLaMA 3 👇
Tweet media one
1
3
8
@SanctumAI
Sanctum
10 months
🧭 Enhanced Navigation: Seamless transitions with our new “Go back” and “Go forward” buttons. ⏰ Chat Grouping: Added “Recent” and “Older” groups for your chat history. 🖼️ Model Icons: Easily identify models with distinct icons in the "New chat" dropdown and chat headers.
Tweet media one
1
1
8
@SanctumAI
Sanctum
8 months
Tweet media one
0
0
7
@SanctumAI
Sanctum
9 months
@bindureddy Check out Sanctum document integration: we're using vector store to get relative chunks of text. Works with .PDFs, more formats coming soon.
0
1
6
@SanctumAI
Sanctum
10 months
📈 Speed Boost: Our switch from .ggml to .gguf model format supercharges the app, expanding our model compatibility game. 🤖 Model Additions: Welcoming Mistral-7B, TinyLlama & Llama-2-13B. 🔖 Bookmarks: Save important chat messages, ensuring they're always within reach.
Tweet media one
1
1
7
@SanctumAI
Sanctum
7 months
We are currently at 12 upvotes on @ProductHunt ! 😼 Lets double it! Can we get to 24? Link below👇
Tweet media one
1
2
7
@SanctumAI
Sanctum
4 months
Check out quants of Mistral-7B-v0.3 on HuggingFace:
Tweet media one
0
2
7
@SanctumAI
Sanctum
4 months
New quants by Sanctum are live on HuggingFace 👇 Coding models ✔️ granite-3b-code-instruct-GGUF ✔️ granite-8b-code-instruct-GGUF ✔️ Codestral-22B-v0.1-GGUF Updated models ✔️ Meta-Llama-3-8B-Instruct-GGUF (support for new tokenizer & EoS token)
1
5
7
@SanctumAI
Sanctum
9 months
We have just been featured on @AITopRank 💜 Help us by upvoting Sanctum here 👇
1
2
7
@SanctumAI
Sanctum
2 months
New model quants available on HuggingFace 👇
Tweet media one
0
2
6
@SanctumAI
Sanctum
10 months
Capping it off, we've given our Sanctum website a thorough makeover. Unlock the power of generative AI in a private environment. Set up your Sanctum today ⤵️
1
1
6
@SanctumAI
Sanctum
10 months
@levelsio We're trying to make it as easy as it can be at Sanctum 🤠
1
1
6
@SanctumAI
Sanctum
5 months
.gguf quants for @MicrosoftAI Phi-3-mini-4k model: (Update with prompt template support for Sanctum coming soon)
0
1
5
@SanctumAI
Sanctum
9 months
⏳ Early Access to New Features With Sanctum Pro, you're always ahead. Experience our latest features and AI model updates first. Are we missing a feature you'd like in Sanctum? Please let us know!
Tweet media one
1
0
5
@SanctumAI
Sanctum
4 months
We were featured on @YahooFinance !💜 "Leader in artificial intelligence integration Sanctum AI made headlines in a recent @VentureBeat article for its innovative approach to solving common cloud-based AI risks faced by today’s businesses." Link below⬇️
1
2
5
@SanctumAI
Sanctum
7 months
With @SanctumAI , your conversations are fully encrypted, guaranteeing they never leave your device. Download today and write with peace of mind. Link below! 👇
Tweet media one
2
1
5
@SanctumAI
Sanctum
11 months
@atari_god "The most important thing is to be aware of what is happening around you and to have the ability to analyze it." -Nelson Mandela
0
0
5
@SanctumAI
Sanctum
5 months
Expect advanced efficiency & precision in specialized tasks. Think role-play, conditionals & the niche brilliance that powered "Chat with AI celebrities on Facebook". LLaMa 3 could be the go-to API for next-gen chatbots. What's your take? Drop your bets below!
1
0
5
@SanctumAI
Sanctum
9 months
Sanctum Pro is available now ✨ Dive into secure PDF interactions, search through encrypted chats, and unlock the full potential of private AI conversations. Download Sanctum and start your 3-day free trial today👇
0
0
5
@SanctumAI
Sanctum
10 months
Tweet media one
0
1
5
@SanctumAI
Sanctum
5 months
🚨 Windows 10 User Alert 🚨 We've identified an issue causing app crashes for some Windows 10 users. A fix is now available! Please download v1.4.2 and reinstall the app to apply the update. Thank you for your patience!
0
1
5
@SanctumAI
Sanctum
1 year
@AGMweb3 @aaronmcdnz @bornreadyvc You narrowed it down 💯 & not just Llama, support for more models coming soon
1
0
3
@SanctumAI
Sanctum
9 months
🔍 Advanced Sanctum Vault Search Find exactly what you need in your encrypted chat history. Search by chat names and messages. Sanctum Pro brings secure, efficient, and local data retrieval to your fingertips.
Tweet media one
1
0
4
@SanctumAI
Sanctum
9 months
🗳️ Secure, 100% Private PDF Chatting Upload and chat with PDFs locally on your device. Summarize, inquire, ask questions, all while keeping your documents private and secure. Stay tuned for more document formats!
Tweet media one
1
0
4
@SanctumAI
Sanctum
5 months
Sanctum v1.4.3 is live 💥 - Fixed bug with chat failing to process subsequent PDF attachments after initial OCR failure. - Added detailed logging to assist with troubleshooting application runtime issues.
0
0
4
@SanctumAI
Sanctum
7 months
@AIandDesign The Mixtral model we currently support has 4096 tokens context. For the next major release we're building a direct integration into HuggingFace so you'll be able to download any GGUF model available there. A lot more improvements for the file reading yet to come.
4
0
4
@SanctumAI
Sanctum
5 months
Andrey's prediction for LLaMa 3 are: ▪️ 2.5M - 3.5M GPU hours for LLaMa 3 ▪️ H100 or B200 GPUs ▪️ $4/hr rate ▪️ Total cost: $10M - $14M
2
0
4
@SanctumAI
Sanctum
5 months
LLaMa 2 70B ran on A100-80GB graphics cards for 1,720,320 GPU hours, at $2/hr, aligning with the estimated ~$3M - $4M spent. So, what's next? 🔮
1
0
3
@SanctumAI
Sanctum
1 year
We are developing an encrypted AI tool that lets you choose which models to use and keeps your conversations and activities private on your local device. Computing power is dependent on which MAC model you have.
@AutismCapital
Autism Capital 🧩
1 year
We would happily pay $100 a month for an AI that actually let us do what we wanted and generate images that are actually funny and brighten our day. Is anyone developing a ChatGPT/Dall-E that isn’t more restrictive than a United Nations Preschool?
74
10
368
0
1
3
@SanctumAI
Sanctum
9 months
@stebco_eth Windows is our next priority
0
0
3
@SanctumAI
Sanctum
1 year
US and European courts are still debating over their AI regulation strategies. The NSA creates the "Artificial Intelligence Security Center" to counter foreign AI threats. AlphaSense secures a jaw-dropping $150M at a soaring $2.5B valuation.
0
0
3
@SanctumAI
Sanctum
10 months
What's the most fascinating prompt you've used with AI? From storytelling to problem-solving, share your favorite AI prompts that sparked innovation!
1
1
3
@SanctumAI
Sanctum
7 months
@Harayokharayo @LordTylerWard Thank you for the support! 💪
0
0
2
@SanctumAI
Sanctum
8 months
@dannypostmaa Run AI locally in a private encrypted environment with Sanctum.
0
0
1
@SanctumAI
Sanctum
11 months
@IFKA_jj You mean the kind that comes from inheriting your parents' hand-me-down furniture and that one weird family recipe? That's the real deal, folks.
0
0
1
@SanctumAI
Sanctum
5 months
0
0
2
@SanctumAI
Sanctum
10 months
@IFKA_jj please try again it seems to be working on our end, otherwise we'll just DM you with the link
0
0
2
@SanctumAI
Sanctum
10 months
@IFKA_jj Did you try to do that with Discord being open?
1
0
1
@SanctumAI
Sanctum
8 months
@digicalidesign We are ecstatic you are loving Sanctum! 🍻
0
0
1
@SanctumAI
Sanctum
5 months
@adelarenal Hi, the Sanctum app itself takes a few hundreds MB of space. More space is needed for AI models and storing chats but you are completely in control of what you're downloading and you can remove old chats as well. Let us know if you have more questions
0
0
1
@SanctumAI
Sanctum
7 months
@pxng0lin Try to keep as little applications open at the same time as possible, that should somewhat help. Also use "Low-Specs" models, those should be less demanding on RAM.
0
0
0