Introducing Clef-omni with full multimodality, plus a faster Clef and a cheaper Clef-flash
Cloudflare has launched the Clef-omni model featuring native multimodal processing for audio, video, images, and text. Additionally, the company has increased Clef inference speeds by 2.0x and reduced pricing for the Clef-flash model.
Verified State Diff
Impact & Verification Analysis
Developers building multimodal AI applications, enterprise users utilizing Cloudflare's AI inference platform, and cost-sensitive AI service providers.
This update improves the competitive positioning of Cloudflare's AI stack by lowering the barrier to entry for multimodal development and reducing the total cost of ownership for high-scale inference workloads.
Full Fact Overview
The introduction of Clef-omni signifies a shift toward unified multimodal architectures, moving away from fragmented pipelines that require separate models for different data modalities. By integrating audio, video, image, and text processing into a single pipeline, Cloudflare reduces latency and architectural complexity for developers building AI-driven applications. The 2.0x speed improvement for the standard Clef model suggests optimizations in the inference engine or model quantization, while the price reduction for Clef-flash indicates a strategic move to capture more market share in the high-volume, cost-sensitive AI inference segment.