Cloudflare ships open-weight Clef-omni with audio/video input, cuts Clef-flash to $0.038/M tokens
Cloudflare Blog · rss · 2026-10-10
Cloudflare released Clef-omni, an open-weight decision model that takes audio, video, image, and text in a single API call — no cascading transcription or frame-extraction pipelines. It's built on a frozen Qwen3-Omni-30B-A3B MoE backbone with LoRA adapters and Brier-score calibration; weights are on Hugging Face.
Speeds: 130 ms median for text decisions, 150 ms for images, 1.5 s for a 21-second video with audio. Benchmarks beat or match TypeSafe's Jev on most metrics (BFCL 98.2, BANKING77 94.8), though When2Call lags (63.3 vs 80.97).
Clef-flash pricing dropped from $0.09 to $0.038/M input tokens — cheaper than Jev — with the hosted context window cut from 64k to 24k (self-hosted weights still support 256k). Clef stays at $0.24/M and got 1.7–2× faster via SGLang serving, merged upstream as PR #42721 for SGLang 0.5.22.
More from Infra
- Full AI video made at home on a single RTX 3090 with Qwen, FLUX, MiniMax and YuE2 — solyarisoftware · 2026-10-10
- Baseten launches Project Beacon with Goodfire for inline open-model safety — baseten · 2026-10-10
- Huawei launches OceanStor M900 context memory storage for hyperscale AI data centers — pstAsiatech · 2026-10-10
- Morgan Stanley: Amazon Trainium3/4 to drive Intel EMIB-M packaging, capacity scaling 8x by 2028 — pstAsiatech · 2026-10-10
- Data centers could become SpaceX's biggest revenue stream, with $1T revenue on the horizon — Dr_Singularity · 2026-10-10
- New gTLD application for .lan shows why internal services need real registered domains — evilsocket · 2026-10-10