DEV Community

#edgeai

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Sipeed MaixStation (AX8860P): 256 TOPS 128GB Edge AI Box
embedsbc profile

Sipeed MaixStation (AX8860P): 256 TOPS 128GB Edge AI Box

#nvidia #edgeai #ai #machinelearning
Comments
8 min read
AAEON PICO-PTH4 Specs: Panther Lake 180 TOPS Pico-ITX SBC
embedsbc profile

AAEON PICO-PTH4 Specs: Panther Lake 180 TOPS Pico-ITX SBC

#ai #edgeai #machinelearning #industrial
Comments
8 min read
The Physical Nightmare of On-Device AI: Why Your LLM Fails on a Custom PCB
maggie-pcb profile

The Physical Nightmare of On-Device AI: Why Your LLM Fails on a Custom PCB

#edgeai #thermalmanagement #pcbdesign #hardwareengineering
Comments
4 min read
One Omni Model, Two GGUF Files: What LFM3-Omni-2B Really Costs in RAM, from the Tensor Headers Up
ginigen_ai_010d9180cbbdb1 profile

One Omni Model, Two GGUF Files: What LFM3-Omni-2B Really Costs in RAM, from the Tensor Headers Up

#ai #machinelearning #edgeai #quantization
Comments
12 min read
Observability for Flutter + Robotics + Edge AI Systems
vmodal_ai profile

Observability for Flutter + Robotics + Edge AI Systems

#flutter #edgeai #robotics
Comments
2 min read
Edge AI Hardware Fragmentation: Why Running an AI Model Everywhere Is Still an Engineering Challenge
y-eager profile

Edge AI Hardware Fragmentation: Why Running an AI Model Everywhere Is Still an Engineering Challenge

#challenge #ai #edgeai #machinelearning
Comments
6 min read
Does 2-Bit Quantization Break On-Device Safety? What an EMNLP 2026 Study Found, and a GGUF Release Gate
ginigen_ai_010d9180cbbdb1 profile

Does 2-Bit Quantization Break On-Device Safety? What an EMNLP 2026 Study Found, and a GGUF Release Gate

#ai #machinelearning #edgeai #security
1
Comments
14 min read
Tri-Hybrid Architecture Deep Dive: Liquid AI's LFM3-1.7B, Sub-Millisecond Convolutions, and 256K Context on Edge Silicon
sharathjapa profile

Tri-Hybrid Architecture Deep Dive: Liquid AI's LFM3-1.7B, Sub-Millisecond Convolutions, and 256K Context on Edge Silicon

#liquidai #lfm3 #architecture #edgeai
Comments
7 min read
Kimi K3 in 8 GB of RAM: Why 28x More Memory Buys Only 1.7x Speed, and the Byte Math That Predicts It
ginigen_ai_010d9180cbbdb1 profile

Kimi K3 in 8 GB of RAM: Why 28x More Memory Buys Only 1.7x Speed, and the Byte Math That Predicts It

#ai #opensource #edgeai #llm
Comments
12 min read
A 763B MoE on a 12 GB GPU: Our Speed Ceiling Versus What Strata Reports, and What Closes the Gap
ginigen_ai_010d9180cbbdb1 profile

A 763B MoE on a 12 GB GPU: Our Speed Ceiling Versus What Strata Reports, and What Closes the Gap

#ai #machinelearning #edgeai #llm
Comments
13 min read
Pinning NPU cores on RK3588 buys you 5% — and costs the big model 10%
anguoyang profile
Anguo

Pinning NPU cores on RK3588 buys you 5% — and costs the big model 10%

#rockchip #rknn #benchmark #edgeai
Comments
2 min read
System design for physical AI: from a defect image to the chamber that drifted
umar_bilal_fd6e1f54398cec profile

System design for physical AI: from a defect image to the chamber that drifted

#ai #architecture #edgeai #ontology
Comments
5 min read
RK3576 vs RK3568: Which Rockchip SoC Is Right For Your Embedded Edge Project?
rocky_gao_66d703529b0f7c6 profile

RK3576 vs RK3568: Which Rockchip SoC Is Right For Your Embedded Edge Project?

#embedded #rockchi #edgeai #linux
2
Comments
3 min read
A 763B MoE at "2 Bits": Where the Bytes Really Go in the GSQ-RCO GGUFs, and Why 100% Recovery Sits Inside the Noise
ginigen_ai_010d9180cbbdb1 profile

A 763B MoE at "2 Bits": Where the Bytes Really Go in the GSQ-RCO GGUFs, and Why 100% Recovery Sits Inside the Noise

#ai #machinelearning #edgeai #quantization
Comments
15 min read
10 RKNN models benchmarked on RK3588 — and the fastest one is cheating onto the CPU
anguoyang profile
Anguo

10 RKNN models benchmarked on RK3588 — and the fastest one is cheating onto the CPU

#rockchip #rknn #benchmark #edgeai
Comments
4 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.