AI Briefing#edge-ai
The latest AI and developer news about #edge-ai, with the original source and a short summary.
The latest AI and developer news about #edge-ai, with the original source and a short summary.
NASA is developing an onboard medical AI assistant using RamaLama to address communication latency issues.
Liquid AI has launched LFM 2.5-230M, an ultra-lightweight parameter model optimized for agentic workflows.
EdgeRazor is a lightweight LLM framework based on mixed-precision quantization and distillation that can be deployed across diverse hardware.
The next-generation YOLO26 model has been released with improved performance through edge device optimization and NMS removal.
Google has unveiled Eloquent, which supports efficient AI model execution on edge devices.
They unveiled **InstinctRazor**, a technology optimized to run MoE models on edge devices.
Google has released Gemma 4 models with QAT technology applied to boost efficiency on mobile and laptop environments.
Google's Gemma 4 12B is a lightweight multimodal model that processes images and audio directly without separate encoders.
Google has unveiled Gemma 4 12B, a new 12B multimodal model that processes vision and audio directly without encoders.
Bonsai Image 4B, an ultra-lightweight image generation model that reduces memory usage to around 1GB through 1-bit and Ternary quantization, has been released.
NVIDIA is extending agentic AI into physical environments on the Jetson platform through JetPack 7.2 and NemoClaw.
The commander of US Special Operations Command emphasized that a very cautious approach is needed when it comes to using AI on the battlefield and integrating it into lethal weapons systems.
Google has unveiled the Coral Board, on-device AI hardware capable of running Gemma 3 locally.
A 65k-scale hybrid router that efficiently switches between local and cloud models has been unveiled.
Sipeed unveiled the K3 RISC-V SBC series, touting 60 TOPS and 30B LLM operation.
MiniCPM-V 4.6, a 1.3B-parameter multimodal model capable of running on smartphones, has been released.
LocalVQE, an ultra-compact audio model that removes real-time echo and noise using about 1 million parameters, has been released.
Open-weight AI models are contributing to improved agricultural productivity and data sovereignty in developing countries such as Kenya.
The AI inference market is fragmenting by use case.
Nvidia has unveiled Nemotron 3 Nano Omni, an open-weight multimodal model.
NVIDIA has unveiled Raw2Insights-US, which estimates speed of sound from raw ultrasound data.
Anker has unveiled the **Thus** AI chip for audio devices.
mii-llm released the 0.4B multilingual models Zagreus/Nesso, trained on 64×A100.
An arXiv paper presents a lightweight English model for low-latency on-device streaming ASR.