Every processed story in chronological order, with the newest coverage first. Filter by tag, source, or score to drill in.
MSA demonstrates that a 109B-parameter model can process 1M-token contexts with 28.4x less attention compute and 14.2x faster prefill, making million-token agentic and code-reasoning workloads substantially more feasible at deployment scale.
Fable 5's availability on AI Gateway brings a model designed for autonomous, multi-day agentic runs — with built-in parallel sub-agent dispatch and stronger code review capabilities — to Vercel's unified inference layer, which offers no-markup provider pricing and BYOK support.