[KongchangAI]
ModelClef-omni

Clef-omni

An open-source multimodal decision model released by Cloudflare, built on Alibaba's Qwen3-Omni-30B-A3B and licensed under Apache 2.0. It supports unified scoring and judgment across text, image, audio, and video modalities within a single API call, primarily targeting content moderation and trust-and-safety use cases.

Timeline (last 90 days)

Oct 13

Cloudflare's Clef-omni model processes text, images, audio, and video in a single API call, with a context window of 24K tokens for the flash tier and 64K tokens for the base tier, and requires Cloudflare Workers AI access.

Unverified50%
Oct 13

Decision-1 and Clef-omni require first aligning with existing latency, token cost, and context window baselines before deciding whether to replace, and it is recommended to evaluate first.

Unverified50%
Oct 12

Clef-omni is capable of scoring and judging content across four modalities—text, images, audio, and video—in a single call.

Unverified50%
Oct 12

Clef-omni charges $0.15 per million input tokens.

Unverified50%
Oct 12

Cloudflare has released a multimodal decision model named Clef-omni, licensed under the Apache-2.0 open-source license.

Unverified50%
Oct 12

Under the Apache-2.0 license, enterprises can use the model on a pay-as-you-go basis through Cloudflare's hosted service, or self-deploy and fine-tune based on the open-source weights.

Unverified50%
Oct 12

The price of Clef-flash is approximately three-quarters lower than that of the standard version.

Unverified50%
Oct 12

Clef-omni is built on Alibaba's Qwen3-Omni-30B-A3B.

Unverified50%

All Facts (8)

Source Articles