Cloudflare releases Clef-omni for unified audio and video decisions
Cloudflare announced Clef-omni, an open-weight decision model that accepts audio and video alongside text and images. Instead of chaining separate speech transcription and video-frame models, it classifies and selects field values from multiple modalities within one request.
Event / data period:
Published: · Updated:
The company says it adapted the Qwen3-Omni-30B-A3B-Instruct foundation, removed speech-generation components and uses two-stage attention routing for schema-bound choices. Cloudflare reports median decision times of 130 ms for text and about 150 ms for images; audio takes a few hundred milliseconds and a 21-second video with sound about 1.5 seconds.
Cloudflare also cut Clef-flash pricing from $0.09 to $0.038 per million input tokens while reducing the hosted context window from 64,000 to 24,000 tokens. Clef-omni starts at $0.15 per million input tokens. Speed and benchmark comparisons come from the company’s tests, not independent verification.
Data
Cloudflare releases Clef-omni for unified audio and video decisions
- · Occurred
Cloudflare releases Clef-omni for unified audio and video decisions
Cloudflare announced Clef-omni, an open-weight decision model that accepts audio and video alongside text and images. Instead of chaining separate speech transcription and video-frame models, it classifies and selects field values from multiple modalities within one request.
Methodology
This event record summarizes the dated statement in the cited primary source. Product or test claims from the source are not presented as independently verified results.
Updated:
Subjects and places
Subject proposals
- Science and technologySuggested
- Preserved editorial category/topic; broad candidate onlyblog.cloudflare.com ↗