Powering the agents: Workers AI now runs large models, starting with Kimi K2.5
ID: 6c5ff9f3-bbb2-5d44-8e4c-43900604a594
STIX ID: report--6c5ff9f3-bbb2-5d44-8e4c-43900604a594
Feed Name: Cloudflare Blog
Cloudflare announces that Workers AI now serves Moonshot AI's Kimi K2.5 model with a 256k context window and features for agentic workloads. The post details optimizations (custom kernels, Infire engine, parallelization, disaggregated prefill), platform improvements (prefix caching, x-session-affinity header, surfaced cached-token metrics), and a revamped asynchronous API for durable inference, aiming to offer high-performance, cost-efficient open-source model hosting for agent deployments.
Your team is not currently subscribed to this feed. You must subscribe to it in order to see this post.