@hacks.grOne PearX startup says its assistant can place calls using an automated voice; another is building AI tools for wealth management.
Investors picked both among five standouts from 16 comp…
https://en.hacks.gr/ependytes-xechorisan-pente-apo-tis-16-etaireies-tis-pearx/
@apexgridtech.bsky.socialBuilding African-language AI is easier than finding the data.
The substrate matters more than the model.
The boring work is where it compounds.
#NigerianAI #OnDeviceAI #AfricaDeepTech #ApexGridTechnologies
@abchaudary.meShould more AI features run on the device? Chrome @developer.chrome.com already gives everyone the tool to check: the Network tab.
What would make you trust an on-device AI feature?
@coxycode.bsky.socialWe build on-device AI for
On-device AI for medical reports
On-device AI for private financial documents
On-device AI for confidential legal contracts
On-device AI for private voice notes
On-device AI for sensitive photos
+91 940-940-8044
[email protected]
@spaisee.bsky.socialOne GPU. A 256K context window. Xing4.0-29B-A4B could reshape local AI, if reliability catches up.
#AI #OnDeviceAI #LLM https://spaisee.com/article/xing4-0-29b-a4b-brings-a-256k-agent-to-one-consumer-gpu
@aidailypost.comJust saw NVIDIA’s DGX Spark with 64GB unified memory crushing 100B‑parameter models on‑device. Thanks to the Grace Blackwell Superchip and DGX OS, local inference is finally a reality for agentic apps. Dive in! #DGXSpark #100BModels #OnDeviceAI
@bithuman.aiOn an iPhone 15, Expression 2 renders at 5.5× real time with the Swift package, and holds 5.1× for 10 minutes.
At 1.0× or more an avatar holds a live conversation, so a character like Yuki, a white arctic fox, can be drawn on the phone itself.
Meet Yuki from the card.
@futuregearai.bsky.socialA new arXiv paper, HARISSA, uses a language model's own hidden states to decide which queries deserve extra compute and which are too risky to answer, helping small local models balance efficiency and safety without sending…
#EdgeAI #OnDeviceAI #LLM #LocalInference
https://arxiv.org/abs/2609.38006
@michel-k-tech.bsky.socialQualcomm's Snapdragon 8 Elite Gen 6 is shifting the Android flagship race toward 2nm thermals and on-device AI instead of quick benchmark wins.
#tech #technology #snapdragon #qualcomm #android #ondeviceai
@futuregearai.bsky.socialA new system called PastForward speeds up on-device GUI agents by reusing fine-grained computational experience from prior tasks as multi-token proposals, reducing the need for full vision-language model inference at every step…
#EdgeAI #GUIagents #OnDeviceAI #VLM
https://arxiv.org/abs/2609.32166
@hacks.grDetectifAI is developing a tool to detect fake voices on the phone itself, without sending audio off-device.
The company says it already checks 100,000+ calls a mo…
https://en.hacks.gr/detectifai-ergaleio-gia-ton-entopismo-pseytikon-fonon-apeytheias-sto-kinito/
@qmshahzad.bsky.socialApple’s new Mac mini and Mac Studio are available with updated Apple silicon and configurations designed for demanding local AI and developer workloads. #Apple #MacStudio #OnDeviceAI
@futuregearai.bsky.socialLocal deployment of LLMs is getting more attention, but energy costs on consumer hardware have been underexplored. A new benchmark tests 18 open-source models (0.5B–7B parameters) on a single RTX 4060ti 16GB using…
#LLM #AIHardware #EnergyEfficiency #OnDeviceAI
https://arxiv.org/abs/2608.00008
@qynque.bsky.socialYou say what you want done — it happens on your machine. Not a suggestion in a chat box, but real work, carried out where your things already live. #QynQue #AIAgents #OwnYourAI #OnDeviceAI #GetThingsDone qynque.com
@qynque.bsky.socialQynQue OS is where an idea meets your desk. Qynix plans it, QuePo builds it, and the work happens right on your own machine. one of the QynQue series #QynQue #QynQueOS #PersonalAI #OnDeviceAI #BuildWithAI qynque.com
@coxycode.bsky.socialWant a babysitter booking app with real privacy — on-device AI, no cloud uploads, no data selling?
📞 +91 940-940-8044
📧 [email protected]
#babysitterapp #childcareapp #dataprivacy #ondeviceai #privacyfirst #parentinghack #appdevelopment #techforparents #ondemandapp #buildyourapp
@spaisee.bsky.socialA 5.9 GB reasoning AI? Bonsai 2 could put private, offline intelligence in your pocket, if battery life and speed hold up.
#AI #OnDeviceAI #Privacy https://spaisee.com/article/prismml-s-bonsai-2-27b-brings-reasoning-ai-closer-to-your-pocket
@apexgridtech.bsky.socialA homegrown model is rising.
The machine in the room is the frontier.
Build for the 8GB laptop in Lagos.
@qynque.bsky.socialYour laptop is the vault. Files never leave, QuePo keeps guard, and Qynix stands beside you. Privacy that lives on your own machine, not someone else's server. one of the QynQue series #QynQue #PrivacyFirst #LocalAI #DataOwnership #OnDeviceAI qynque.com
@apexgridtech.bsky.socialEdge AI is not a luxury. It is the substrate.
The machine in the room is not waiting for permission to compute.
We are building the intelligence layer for the 8GB laptop in Lagos.