Hardware

7 articles

Hardware

Clawpit

340M-parameter GLiNER2.5-Decide runs on phone GPU with numerical parity to reference model

Fastino researchers have released a LiteRT build of GLiNER2.5-Decide, a DeBERTa-v3-large derivative that performs zero-shot classification with arbitr…

HardwareAuto-updating

Liquid AI releases experimental drafter accelerating vision-language decoding up to 3.13x on Apple silicon
Liquid AI has announced LFM2.5-VL-3B-DSpark, an experimental drafter model that sits atop its LFM2.5-VL-3B vision-language model and promises up to 3.

Hardware26.09.2026New
Fireworks cuts RL delta size in half, makes cross-region training practical
The bottleneck in frontier model training isn't the GPU supply — it's the network. Fireworks is launching ARCv3, a new compressor for reinforcement-le

Hardware24.09.2026New
Anthropic blocks Chinese hardware and Trainium in new Claude Opus 5.5
An X user, xlr8harder, reports that Anthropic’s new model, Claude Opus 5.5, refuses requests related to development on Chinese hardware. According to

Hardware23.09.2026New
Pennsylvania data centers: resident opposition halts massive investments
Leading AI labs are convinced that scaling compute is the path to better models, and they are backing that belief with plans for trillions of dollars

Hardware22.09.2026
California makes AI data centers pay for grid upgrades and disclose water use
Governor Gavin Newsom has signed a package of seven bills designed to stop AI data centers from passing electricity and water costs on to residents. T

Hardware22.09.2026
VLLM offloads video decoding to NVDEC, doubles throughput on 8×H100
The vLLM inference library has integrated PyNvVideoCodec and moved video decoding from the CPU to NVDEC, Nvidia's hardware decoder, removing a bottlen

Hardware22.09.2026