open source AI - Page 16
Stories from open source AI come first, newest first, then stories from connected topics (Qwen, Open Source, open weights, AI Models, Hugging Face, AI), each labelled with where it comes from. This view is not indexed by search engines.
Latest
Tensormesh Launches Software to Reuse AI Models’ “KV Cache” Across Enterprise Servers. Here Is Why That Matters
Tensormesh, maker of LMCache, has launched software to store and reuse KV cache across enterprise AI servers. What a KV cache is, and why reuse matters.
UltraBERT Explained: Open 32K-Context Encoders, and Why “How Much You Hide” Decides What a Model Is Good At
DMIS Lab's UltraBERT is a family of open 32K-context encoders. Explainer: what encoders do, and why the masking finding matters for search and RAG.
Inworld Launches Inworld Inference, an API It Says Is Built for Consumer AI Apps
Inworld has opened Inworld Inference, an API for running open AI models in consumer apps. It says it is faster and cheaper than the median provider.
Mozilla Launches Firefox Enterprise, a Business Browser Companies Can Manage on Their Own Servers
Mozilla has introduced Firefox Enterprise, a business browser run from a console organisations host themselves. Sales start in four European countries.
Hugging Face Launches Llama Windows, a Free App That Runs Open AI Models From Your Windows 11 Tray
Hugging Face's Llama Windows runs open AI models on your Windows 11 PC from the tray, built on llama.cpp. Free, open source, no terminal or admin rights.
Liquid AI Releases LFM2.5-24B-A2B, an Open-Weight Model for Running AI Agents on Your Own Device
Liquid AI's open-weight LFM2.5-24B-A2B has 24B parameters but uses only 2B at a time, built to run AI agents fast and privately on your own device.
Docker Hub Flaw Handed Users’ Access Tokens to New Owners When Accounts Became Organisations; Docker Says It Is Fixed
A Docker Hub flaw, CVE-2026-105445, handed users' access tokens to new owners when accounts became organisations. Docker says it is fixed and saw no abuse.
NVIDIA Makes Its Full Run:ai Platform Open Source as Part of NVIDIA DSX AI Factory
NVIDIA has released its full Run:ai GPU orchestration platform as open source, in four parts usable separately. Eight partners are integrating it.
IBM Research Says Open AI Agents Sped Up a Quantum Decoder Chip Design 52 Times for About $26
IBM Research says open AI models served through llm-d sped up a quantum error-correction decoder design 52x in synthesis estimates for about $26.
IBM Moves Its Spyre AI Chip’s Compiler Onto PyTorch’s Open-Source Inductor
IBM's Spyre accelerator now compiles through PyTorch's open-source Inductor. IBM says the new compiler code is about 7x smaller than the one it replaced.
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.