Trending

#edgeAI

Latest posts tagged with #edgeAI on Bluesky

Posts tagged #edgeAI

Post image

GuardAI meets RECITALS!
We explored synergies between edge AI security & privacy-preserving tech. Key topics: technical knowledge exchange, joint dissemination, cross-project integration.
Thanks RECITALS for the insightful dialogue & shared Trustworthy AI goals! #TrustworthyAI #EdgeAI #EUInnovation

0 0 0 0
Post image Post image Post image

GuardAI at "Driving Trustworthy AI through Design and Regulation"

We engaged policymakers, researchers & industry on secure-by-design AI, regulatory compliance & deployment.
Great chat with Sebastian Hallensleben (JTC 21 Chair) on AI standards.
#TrustworthyAI #AIRegulation #EdgeAI #AIbyDesign

0 0 0 0
Post image

Auf dem Mobile World Congress in Barcelona präsentierte Supermicro eine erweiterte Unterstützung für Infrastrukturlösungen, die Plattformen für souveräne KI sowie KI-RAN (Künstliche Intelligenz in Funkzugangsnetzen) antreiben. #5G #AIinTelecom #EdgeAI #MWC #SovereignAI
https://www.sysbus.eu/?p=31037

0 0 0 0

...we are the data center. 🚀📱 #DecentralizedAI #DePIN #Web3
-
Modern Androids pack 45+ TOPS of NPU compute. Instead of draining battery, a native app using ExecuTorch and NNAPI runs 4-bit quantized LLMs natively, turning smartphones into a massive, parallelized compute swarm. #EdgeAI #AndroidDev ...

1 0 2 0
Preview
Microsoft brings sovereign edge AI to Industry 4.0 private networks Bringing sovereign edge AI to remote sites, a collaboration between Microsoft and Armada accelerates private networks for Industry 4.0.

Microsoft is bringing sovereign edge AI directly to disconnected environments like factory floors, remote mining sites, and offshore rigs. #microsoft #private5g #edgeai #iiot #telecoms #iot #ai #5g #digitaltransformation #industry40 #technology

0 0 0 0
Preview
Agricultural machinery achieves 81% harvest success via edge IoT Fusing AI with edge IoT sensors in agricultural machinery has pushed automated harvest success to 81 percent, altering industrial economics.

Researcher have fused AI with edge IoT sensors, giving agricultural machinery the ability to evaluate the physical difficulty of picking a crop before initiating mechanical action. #iiot #edgeai #agriculture #farming #iot #automation #smartfarming #agtech #robotics #ai #technology

0 0 0 0
Video

AI‑powered farming is becoming real.
Robots using vision AI to remove weeds without chemicals show how physical AI can transform agriculture sustainably.

#AI #Robotics #AgTech #Sustainability #EdgeAI

1 0 0 0
Preview
【2026年版】エッジAI実装完全ガイド ─ デバイスで動くAI推論の最新手法と実践

2026年、AI推論のトレンドがクラウドからエッジへ劇的にシフトしてますね。なんと推論の55%がデバイス内で完結する時代に。

特に「Qwen 2.5-0.5B」のようなSLMが512MBのVRAMで日本語対応しているのは衝撃。NPU活用でスマホ開発も激変しそう。

みんなはエッジAI、何で動かしてる?🤔

#AI #EdgeAI #LocalLLM #Python

https://zenn.dev/ai_nexus/articles/edge-ai-2026-guide

2 0 0 0
Preview
Autonomous Robotics Is Stuck at 99%, And That Last 1% Is Holding Back Scale - KoreaTechDesk | Korean Startup and Technology News Autonomous robotics stalls at 99% reliability. A veteran CEO explains edge-case failures, AI limits, and what this means for scaling and Korea’s OEMs.

Autonomous robots can hit 99% reliability.

That last 1% is the stopping scale.

Edge cases, real-world chaos, and system integration are still breaking deployments—especially in high-adoption markets like Korea. ⚡️

#Robotics #DeepTech #EdgeAI #startups
koreatechdesk.com/autonomous-r...

0 0 0 0
Post image

most amazing on-device model now,

did you even try it, lol ?

#ai #gemma #edgeai #chatgpt

2 0 0 0
Preview
Google Previews Gemini Nano 4 for Android AICore Google has opened a developer preview for Gemini Nano 4, its next on-device AI model for Android, promising 4x faster inference and 60% lower battery use.

winbuzzer.com/2026/04/04/g...

Google Previews Gemini Nano 4 for Android AICore

#AI #Google #GoogleGemini #GeminiNano #Android #GenAI #AIModels #OnDeviceAI #EdgeAI #Smartphones #BigTech #Alphabet #OpenSourceAI

0 0 0 0
Post image

#EdgeAI in #PredictiveMaintenance: A Look at Workflows, Challenges and Outcomes www.industryweek.com/technology-a...

0 0 0 0

That's why I tagged it #edgeLLM / #edgeAI.

I expect to see #RISC-V #vectorISA get more sophisticated with #TurboQuant or at least #keyvaluecache, #takum & #1bit, maybe much more sophisticated lookups for new #quintic geode & #adelic approaches relying on #tribonnaci etc.

M could just add that.

0 0 0 0
Preview
Running AI on a Raspberry Pi, Part 2: Running AI on a Pi in Under 5 Minutes -- Virtualization Review Running an LLM on a Raspberry Pi with Ollama is easy to set up in minutes, but practical performance depends heavily on choosing a smaller model tuned for low-power hardware.

Tom Fenton shows how running AI on a Raspberry Pi is now practical, with Ollama enabling local LLMs in minutes and lightweight models improving performance.

See how to run AI on a Pi in minutes: https://ow.ly/y1yq50YBJjh

#EdgeAI #RaspberryPi #AI

0 0 0 0
Preview
Running AI on a Raspberry Pi, Part 1: Overview -- Virtualization Review This first article in a series explains the core AI concepts behind running LLM and RAG workloads on a Raspberry Pi, including why local AI is useful and what tradeoffs to expect.

Running AI on a Raspberry Pi is now possible with smaller LLMs and RAG, enabling local, private and offline AI workflows, as Tom Fenton explains.

See how AI can run on lightweight hardware: https://ow.ly/5r6n50YAVta

#AI #EdgeAI #RaspberryPi

1 0 0 0
Post image

I spoke with several ECE professors at #ECEDHA 2026 about how they’re integrating #AI into the curriculum. Everyone agrees it’s essential, but no one agrees on the best way to teach it. Check out my video interviews to learn more:

shawnhymel.com/3246/ai-in-e...

#engineer #education #edgeAI

2 0 0 0
Preview
Introducing The P100 COMx6 for Rugged Edge AI | SolidRun SolidRun's new  COM Express Type 6 module family based on AMD Ryzen™ AI Embedded P100 processors, designed for developers building rugged edge AI systems.

Turns out running 50 TOPS of AI inference on a module rated for -40 to +85 degrees is now a thing. Edge AI for rail, robotics, and industrial systems just got a serious hardware option. www.solid-run.com/blog/article... #EdgeAI

0 0 0 0
Post image

EDGEAI is keeping utility meter batteries alive for 8 years with the help of BrainChip's ultra low power Akida 2 IP.
Event-based compute. No battery compromise.

Full announcement: https://bit.ly/4185gGS

#EdgeAI #SmartMetering #IoT #Akida2 #NeuromorphicAI

0 0 0 0

8/9
Scenario:
You’re on a train with no signal.
Your AI is still planning your week, organizing tasks, suggesting priorities.

This is already possible today.
#EdgeAI #AI #OfflineAI #YourAI

0 0 1 0

7/9
Reality check:
Local models are still weaker than cloud ones.
But for daily workflows, they are already enough.
And you stay fully independent.
#AI #EdgeAI #PersonalAI #YourAI

0 0 1 0

5/9
Pros:
Privacy
Works without internet
Fast responses

Cons:
Heats up on heavy tasks

Still worth it for full independence.
#EdgeAI #PrivacyTech #YourAI

0 0 1 0

4/9
Setup in ~15 minutes:

MLC LLM or Ollama mobile
~8B model
Base prompt:

“You are my daily agent. Plan, remind, analyze proactively.”
#AgenticAI #EdgeAI #YourAI

0 0 1 0

2/9
Edge AI means local: no internet, fast, private.
No cloud dependency, no passive data collection.
Models like Llama and Gemma now run directly on phones.
#EdgeAI #Privacy #PersonalAI #YourAI

0 0 1 0
Post image

1/9
My phone offline already plans my day, reminds meds, even reviews code.
Edge + agentic AI in 2026 = a second brain in your pocket.
Want it? Here’s how to set it up.
#AI #AgenticAI #EdgeAI #YourAI

0 0 1 0
Post image

🗣️Prior to the RK3688 launch, ArmSoM plans to develop two new products in 2026.Something new is coming to the Sige family:Sige6.
Our first @Allwinner board powered by A733.
And we want YOU to help define it.
Drop your ideas below !!!!!
#edgeAI #embedded
www.allwinnertech.com/index.php?c=...

1 0 1 0
Video

"Think a driver is fine because they’re not yawning? Think again. DriverCheck by CSEM spots subtle physiological warning signs with contactless, privacy-first AI. See how it works: bit.ly/47lUtfQ
#RoadSafety #EdgeAI"

2 0 1 0
Preview
The Era of Autonomous AI Systems Artificial Intelligence (AI) has evolved rapidly over the past decade. What once started as simple automation tools and machine learning algorithms has now grown into highly advanced systems capabl...

The Era of Autonomous AI Systems
www.ekascloud.com/our-blog/the...
#AutonomousAI #ArtificialIntelligence #AI #MachineLearning #FutureTech #CloudComputing #IntelligentSystems #TechInnovation #AIRevolution #Automation #DigitalTransformation #EdgeAI #AIWorld #NextGenTech #AIApplications

1 0 0 0
Preview
Lattice Semiconductor Joins NVIDIA Halos AI Lab Lattice (NASDAQ: LSCC) joined NVIDIA's Halos AI inspection lab on Mar 29, 2026; design-win conversions typically take 6–24 months, per industry procurement cycles.

Lattice Semiconductor Joins NVIDIA Halos AI Lab: Lattice (NASDAQ: LSCC) joined NVIDIA's Halos AI inspection lab on Mar 29, 2026; design-win conversions typically take 6–24 months, per industry procurement cycles. 👈 Read full analysis #LatticeSemiconductor #NVIDIA #HalosAI #AIInspection #EdgeAI

0 0 0 0
The Developer's Guide to Qwen 3.5 0.8B: A Hands-On Tutorial for Local Deployment Learn how to locally deploy Qwen 3.5 0.8B, an efficient open-source Small Language Model (SLM) for edge AI, multimodal tasks, and on-device inference.

✨ 0.8B parameters.
🧠 262K context window.
👁️ Native multimodal vision.
💻 Runs on your MacBook.

If you aren't playing around with local AI yet, Qwen 3.5 0.8B is the perfect place to start. Read my new blog post here: tinyweights.dev/posts/develo...

#EdgeAI #Qwen #LocalAI

1 0 0 0
Preview
Cutting Costs with Edge AI: The Rise of SLMs The generative AI boom taught enterprises a hard lesson: intelligence is easy to demo and brutally expensive to scale. The same architectures that make frontier models impressive—huge parameter counts, massive context windows, GPU-heavy serving stacks—also create a cost structure that looks less like “software” and more like “utilities.” And in the real world of procurement cycles, latency SLAs, data residency, and FinOps scrutiny, utility-style cost curves don’t win for long.

The future of AI isn’t bigger models—it’s smarter placement. Small, edge‑optimized models are becoming the enterprise default.
#AIArchitecture #EdgeAI #CloudDailywire #EnterpriseAI

0 0 0 0