AI Judge changed title after evaluation, original title: "Jon Saad-Falcon releases OpenJarvisAI v1.0 to run local, low-power AI agents on user devices"
The desktop application includes eight prebuilt workflow agents
Simran Arora#1242
The dominant story in AI has been the growing cloud: bigger clusters, larger models, more gigawatts.
We believe the future is in the opposite direction: on-device inference, smaller models, watts instead of gigawatts.
Today we're releasing @OpenJarvisAI v1.0: a personal AI assistant that lives, learns, and works on your device.
Positive users congratulated the OpenJarvisAI V1.0 launch for its on-device personal AI assistance while negative users dismissed it as another generic chat agent and raised privacy and model-bloat concerns.
No Digg Deeper questions have been answered for this story yet.
Local models deserve way more attention because they are getting incredibly good! Check out @OpenJarvisAI, a personal AI assistant that lives locally on your device and continually becomes more useful to you.
Give it a try! It's a lot of fun! (already at 5.5k stars!! ⭐️)
The dominant story in AI has been the growing cloud: bigger clusters, larger models, more gigawatts.
We believe the future is in the opposite direction: on-device inference, smaller models, watts instead of gigawatts.
Today we're releasing @OpenJarvisAI v1.0: a personal AI assistant that lives, learns, and works on your device.
Huge congrats to @jonsaadfalcon, @Avanika15, @Azaliamirh and the @HazyResearch team on @OpenJarvisAI — out today.
For two years, they've been making the case that AI inference belongs on hardware people already own, not just in megawatt data centers. Excited to support the Intelligence per Watt line of work.
Read more on their blog: https://hazyresearch.stanford.edu/blog/2026-05-15-minions-to-openjarvis-retrospective
The dominant story in AI has been the growing cloud: bigger clusters, larger models, more gigawatts.
We believe the future is in the opposite direction: on-device inference, smaller models, watts instead of gigawatts.
Today we're releasing @OpenJarvisAI v1.0: a personal AI assistant that lives, learns, and works on your device.
I've been excited about the potential of local-first software for nearly a decade; I now believe genAI apps might be the perfect use case for it. Local-first AI apps will be faster, more private, and more efficient. In other words, usable. Super stoked for this release.
Super excited to release OpenJarvis!
Start using it for the most functional, complete, secure, and ipw-efficient personal agent experience!
Cool stuff! DC inference is supply bound - makes sense to offload intelligence locally when you can!
The dominant story in AI has been the growing cloud: bigger clusters, larger models, more gigawatts.
We believe the future is in the opposite direction: on-device inference, smaller models, watts instead of gigawatts.
Today we're releasing @OpenJarvisAI v1.0: a personal AI assistant that lives, learns, and works on your device.
So excited for this bold vision of small, personalized, on-device AI that everyone can own themselves!
Excited to support the multi-objective evaluations that chart this new quality/latency/efficiency/cost/privacy Pareto frontier via @SnorkelAI Open Benchmarks Grants.
The dominant story in AI has been the growing cloud: bigger clusters, larger models, more gigawatts.
We believe the future is in the opposite direction: on-device inference, smaller models, watts instead of gigawatts.
Today we're releasing @OpenJarvisAI v1.0: a personal AI assistant that lives, learns, and works on your device.
@JonSaadFalcon @OpenJarvisAI this is epic. was just talking about something like this today with @sudojbird
@JonSaadFalcon @OpenJarvisAI Congrats @JonSaadFalcon and @BhathalTanvir0 !!! Great work
@JonSaadFalcon @OpenJarvisAI on-device inference is the right bet. what's the license though? actually open or 'open'?
@AI_WarriorNQ @OpenJarvisAI It's under Apache 2.0
OpenJarvis is a personal AI assistant that runs entirely on your devices.
We ship eight prebuilt agents so you can be productive on day one: email triage, morning briefings, document QA, deep research, scheduled automations, code review, and more. Drop them onto your laptop and they'll get to work!
Use them from the CLI, the browser, a native desktop app, or 26+ messaging channels (iMessage, WhatsApp, Telegram, Slack, ...).
OpenJarvis decomposes a personal AI system into five composable primitives:
🧠 Intelligence: on-device LMs (Qwen 3.6, Gemma 4, Kimi 2.5, GPT-OSS, IBM Granite, ...) ⚙️ Engine: inference runtimes (@ollama, @ggerganov's llama.cpp, @vllm_project, @sgl_project, @exolabs, Apple FM, ...) 🤖 Agents: agentic frameworks (ReAct, @OpenHandsDev, ...) 🔧 Tools & Memory: MCP, A2A, semantic storage, 26+ channels, telemetry 📚 Learning: self-improvement from local traces, using gradient-based updates (SFT, GRPO) and prompt optimizers (@DSPyOSS, @gepa_ai)
Each slot is independent. Swap any one without rewriting the rest!
We believe the future of AI is local by default and hybrid by design, unifying local and cloud compute.
Cloud brings frontier capability and elastic scale. Local brings privacy, low latency, and energy efficiency. Used together, you get the best of each and mitigate the limits of both.
OpenJarvis v1.0 supports existing local-cloud collaboration protocols (Minions, Advisor Models, Conductor, ToolOrchestra, SkillOrchestra, and others) and provides the substrate for developing new ones.
We see local-cloud LM systems as a first-class research area, not a footnote. The full arc is in our retrospective:
📝 Blog: https://hazyresearch.stanford.edu/blog/2026-05-15-minions-to-openjarvis-retrospective
We agree 100% The future of AI is in the POND: Personal & Private On-device & On-premise Nodal & Networked Distributed & Decentralized
Huge thanks to the team that made this possible: @robbymanihani, @BhathalTanvir0, @krypticmouse, @akenginorhun, @gabekbo, @andrewparkk, Matthew Hart, @CaiaCostello, Chuan Li (@LambdaAPI), @hazyresearch, and @Azaliamirh.
And thank you to @StanfordAILab, @stanfordnlp, @HazyResearch, @ollama (s/o @jmorgan and Michael Chiang), @IBMResearch, @LaudeInstitute, @huggingface, @googlecloud, @LambdaAPI, @SnorkelAI, and @StanfordData (Marlowe) for the compute and support.
Try OpenJarvis on your own hardware. We'd love to hear what you build with it:
git clone https://github.com/open-jarvis/OpenJarvis.git
cd OpenJarvis
uv sync
jarvis init
💻 Code: http://github.com/open-jarvis/OpenJarvis
📖 Paper: https://arxiv.org/abs/2605.17172
💬 Discord: https://discord.gg/6ZtCB94h5p
🐦 Follow @OpenJarvisAI for updates
@JonSaadFalcon @OpenJarvisAI Very cool
@JonSaadFalcon @OpenJarvisAI Incredible!
@JonSaadFalcon @OpenJarvisAI Congrats @JonSaadFalcon @Avanika15!
@JonSaadFalcon @OpenJarvisAI I’m a huge believer in on-device AI. Will try and happy to help in any way. Congrats!
@JonSaadFalcon @OpenJarvisAI can you help me run it on my tamagotchi