Cloudflare launches multimodal model clairic-flash, trending on Hugging Face

Cloudflare · hf · 2026-11-21

Cloudflare has released clairic-flash on Hugging Face, an image-text-to-text multimodal model now trending on the platform.

Tags suggest it is built on a Qwen3.5 architecture, ships in transformers/safetensors format, and targets structured output and information extraction workloads with mixed image-text input. Cloudflare has not yet published details like parameter count, license, or benchmark results — notable as it's the company's first self-branded model appearance on HF trending.

Original post →

More from Infra

Infra channel →