llm

The Bytecode Alliance patches 12 Wasmtime flaws, two critical, found during a three-week LLM-assisted security sprint by Mozilla, UCSD, Akamai, and F5.

A Caltech spinout backed by Khosla Ventures has released a family of fully 1-bit language models under Apache 2.0 that compress an 8-billion-parameter model to 1.15 GB while matching full-precision rivals on standard benchmarks.

Mistral releases Small 4, a 119B-parameter open-weight model under Apache 2.0, and Forge, an enterprise platform for training custom models on proprietary data, as open-weight AI closes to within three months of frontier systems.

Alibaba releases Qwen3.5, a 397-billion-parameter open-weight model that activates only 17B parameters per pass, beats rivals on instruction-following benchmarks, and ships at 1/18th the cost of Gemini 3 Pro.

A RedMonk study of 32 organizations reveals a fragmented policy landscape as projects from cURL to the Linux kernel draw competing lines on AI code.

Anthropic releases its most capable model yet, featuring a 1M-token context window, parallel agent coordination, and security research that uncovered over 500 previously unknown vulnerabilities in open-source software.