Cloudflare released Clef and Clef-flash, two decision models for AI agents, on October 2, 2026, saying it enables automated decision-making without human intervention. It is the company's first major AI model update since its July 2026 announcement about giving website owners control over AI bot access.
Cloudflare reported median response times of 39 milliseconds for Clef-flash and 209 milliseconds for Clef, measured across 43 benchmarks. That compares with just over 524 milliseconds for TypeSafe AI's Jev.
Clef is built on Qwen3.8-27B and Clef-flash on Qwen3.5-9B, and targets use cases like customer support routing and threat intelligence classification. Availability begins with the release of the models, initially for developers and enterprise customers.
"A human does not necessarily need to be in the loop for agentic decisions anymore," said Cloudflare. The models allow agents to programmatically gather context, make decisions, and take actions on tasks, or defer to a human when needed.
The announcement follows TypeSafe AI's release of Jev in mid-September 2026. Cloudflare frames the significance as a direct response to Jev, offering faster and more accurate decision-making capabilities.
Cloudflare did not say how it will handle edge cases or ambiguous inputs, and raised the question of whether the models can fully replace human oversight in complex scenarios. The company said it plans to use Clef internally for abuse report reviews and support request sorting.
Source: thedecoder
