Tab leverages the memories and activity you already have built up in AnythingLLM - make it a better suggestion engine the more you use it and write.
Download today for Mac and Windows:
anythingllm.com/desktop
AnythingLLM now ships with an OS-wide writing assistant called Magic Tab. Simply clicking on an input can make a contextual suggestion based on what you are looking at as well as contextual typeahead suggestions.
Free to use, powered with on-device models and seamlessly works with our tools, agents, memories, and more for Mac and Windows devices.
This is a better way to dictate. Download AnythingLLM for desktop today at anythingllm.com/desktop
"Intelligent Dictation" is live in @AnythingLLM. Now your local models clean up your words and paste them at your mouse cursor, but also can even see your open windows and give you better outputs based on what you're doing and what you said.
AnythingLLM is an app layer for local models. We believe that a local model can feel like a cloud model with the right tools and guardrails.
We believe we can bring 80% of your daily inference tasks to the computer you are already on.
anythingllm.com
The Bonsai models scale from mobile to Desktop (1.7B, 4B, and 8B) so you can get dense intelligence on any hardware.
You can find the full collection (including MLX) here.
huggingface.co/collections/pr…
Today, we demoed @PrismML's Bonsai 8B model in @AnythingLLM desktop to see if the 1-bit mode arch is the revolution local AI needs.
Turns out, this is not an April fools joke. 1.1GB memory footprint for a whole 8B model with high accuracy
Over the weeked we passed by 50,000 GitHub stars on AnythingLLM!
We are so lucky to have so many people inspired, using, forking, and contributing to our vision of an LLM agnostic tool suite you can run locally or for your organization.
⭐️ github.com/Mintplex-Labs/…
If you have specific models, benchmarks, and other details you want us to try out, let me know, and we can publish those in a follow-up video.
GA for these devices is published to be on Wednesday, Oct 15th, so you can let us try out your builds prior to get ideas.
128GB unified memory, 20-core ARM CPU, 4.1TB of storage, GB10 Blackwell superchip, and more. If you are interested in local AI inference in a single self-contained compute device, our overview video is worth checking out @NVIDIA_AI_PC 👉 youtu.be/zs-J9sKxvoM
We are a lucky partner to bench the new @NVIDIA DGX Spark before it came out. We build AI tools literally every day, so we were eager to get our hands on this to kick the tires.
Accelerate local LLMs and advanced AI workflows in @AnythingLLM with NVIDIA RTX. 🖥️💬
⚡ Fast inference w/ NVIDIA NIM Microservices
📁 Local privacy
🔁 Easy to use UI
Learn more in the #RTXAIGarage blog 👉 nvda.ws/4kA8J9r
NVIDIA & Microsoft Advance Development on RTX AI PCs.🤝
TensorRT has been reimagined for RTX. Available via Microsoft’s new inference stack, Windows ML, they streamline development and unlock peak performance.
Explore #MSBuild & #COMPUTEX2025 news 👉 nvda.ws/4k2tBWr
AI at your fingertips. ✨
NVIDIA NIM microservices arrive on RTX AI PCs & workstations making AI tool creation easier than ever.
Plus, Project G-Assist System Assistant expands PC AI abilities with a custom plugin builder.
#RTXAIGarage: nvda.ws/4iZLlRn
⚔️ 𝗠𝗲𝗲𝘁 𝘁𝗵𝗲 𝗞𝗻𝗶𝗴𝗵𝘁𝘀 𝗮𝘁 𝗟𝗮𝗻𝗰𝗲𝗹𝗼𝘁 𝗥𝗼𝘂𝗻𝗱 𝗧𝗮𝗯𝗹𝗲 - 𝗧𝗶𝗺 ⚔️
@tcarambat is a fellow @ycombinator alumni, and has never been shy about his enthusiasm for LanceDB. Being a loyal ambassador, his team ships @AnythingLLM with LanceDB as a default for a
⚔️ 𝗠𝗲𝗲𝘁 𝘁𝗵𝗲 𝗞𝗻𝗶𝗴𝗵𝘁𝘀 𝗮𝘁 𝗟𝗮𝗻𝗰𝗲𝗹𝗼𝘁 𝗥𝗼𝘂𝗻𝗱 𝗧𝗮𝗯𝗹𝗲 - 𝗧𝗶𝗺 ⚔️
@tcarambat is a fellow @ycombinator alumni, and has never been shy about his enthusiasm for LanceDB. Being a loyal ambassador, his team ships @AnythingLLM with LanceDB as a default for a seamless end-user experience!
🛡️ Becoming a Legendary Lancelot 🛡️
Introducing the Lancelot Round Table - where #opensource heroes transform AI's future. With 2 million monthly downloads, we are seeking bold innovators to reshape how the world thinks about AI infrastructure. We're not just building a database, but a project that will change everything.
blog.lancedb.com/the-lancelot-c…
"Now, with generative AI and RTX AI PCs, anyone can be a developer. A new wave of low-code and no-code tools, such as AnythingLLM, [...], enable enthusiasts to use AI models in complex workflows via simple graphical user interfaces."
New wave of on-device AI tools 🌊
AI that's NIMble.🖥️
We're accelerating generative AI with NVIDIA NIM Microservices, AI Blueprints and NIM-ready RTX AI PCs to help developers and enthusiasts build AI agents, like Project R2X, and enhance creative workflows on PC.
Learn more👉 nvda.ws/4aefadJ
1 Followers 35 FollowingGroup Exercise Instructor, Les Mills Certified BodyPump, LM Core, RPM instructor. Software developer. Amateur short order cook. MS Chemistry UAB, MBA SMU.
908 Followers 3K FollowingAt the intersection of culture and applied AI. Working on transformation beyond the tech bubble. bootstrapping Cultural Policy Lab
2K Followers 719 FollowingLambos, Llamas & good vibes. Join this coin in the making:
Info: https://t.co/2kYz7UB9u6
TG: https://t.co/pUAUenMYac
Discord: https://t.co/htNGeWIxno
25 Followers 168 FollowingMortgage Guy. Not Programmer.
Can AI agents make us more money and give us more time?
I have no idea what I'm doing either. Let's find out.
5K Followers 66 FollowingThe multimodal lakehouse for AI, accelerating large-scale data curation and feature engineering so teams can build better models faster.