Public story · 2026-03-20 · source-backed
Cloudflare Workers AI Ships Large Model Support with Kimi K2.5
Story
Cloudflare announced Workers AI now supports large-scale model inference, starting with Kimi K2.5 (256K context, vision, multi-turn tool calling). Internal testing: switching a 7-billion-token/day security code review agent from a proprietary mid-tier model to Kimi K2.5 achieved a 77% cost reduction. New prefix caching with session affinity headers and redesigned async APIs target the cold-token-cost bottleneck that kills agent economics at volume. Cloudflare just shifted from CDN to full-stack AI inference layer.
Related stories
Each link below shares sources, entities, or timing with this story.
Shared entities / Same source / Shared topic / What happened next
Kimi K2.5 Now on Cloudflare Workers AI — Frontier Open Model on Edge Infrastructure with 77% Cost Savings
Both cover Cloudflare, Kimi K2, Workers AI; cite the same source (Cloudflare announced); overlapping topics (agent, cloudflare, kimi, model).
Shared entities / Same source domain / Shared topic / What happened next
Cloudflare Agents Week: Email, Git Artifacts, 22% Lossless LLM Compression, and Agent Memory
Both cover CDN, Cloudflare; reported by the same outlet (blog.cloudflare.com); overlapping topics (agent, cloudflare, model, worker).
Cloudflare's Two-Tool MCP Pattern Cuts 1.17M Tokens to 1,000
Both cover APIs, Cloudflare; reported by the same outlet (blog.cloudflare.com); overlapping topics (agent, cloudflare, code).
Shared entities / Shared topic / What happened next / Tension
Computer-Use Agents Cost 45x More Than APIs. The Numbers Are Brutal.
Both cover APIs, Internal; overlapping topics (agent, apis, model); picks up the APIs thread on 2026-05-06.
Kimi K2.6 Beats GPT-5.4 on SWE-Bench Pro. It's Open-Weight. It Costs 88% Less.
Both cover APIs, Kimi K2; overlapping topics (agent, kimi, model); picks up the APIs thread on 2026-05-05.
Shared entities / Same source domain / Shared topic / What happened next
Cloudflare turns HTTP 402 into a product
Both cover CDN, Cloudflare; reported by the same outlet (blog.cloudflare.com); overlapping topics (agent, cloudflare).
Shared entity: Cloudflare / Same source domain / Shared topic / What happened next
Cloudflare Dynamic Workers: V8 Isolate Sandboxing for AI Agent Code Execution, 100x Faster Than Containers
Both cover Cloudflare; reported by the same outlet (blog.cloudflare.com); overlapping topics (agent, cloudflare, code, worker).
Shared entities / Shared topic / What happened next
Kimi K2.7-Code: a 1T-param open coding model with benchmarks nobody can check
Both cover APIs, Kimi K2; overlapping topics (code, kimi, model); picks up the APIs thread on 2026-06-14.
Source trail
Entities
Provenance
- Canonical issue
- Ramsay Research Agent
- AI generated
- no
- Story unit
- 2026-03-20-cloudflare-workers-ai-ships-large-model-support-with-kimi-k2-5
- Labels
- source-backed, canonical briefing excerpt