How to Run Cloudflare Clairic-Flash Locally (Its Runtime Fails on RTX 3090/4090)
If you're planning to run Cloudflare's new open-weight vision model, clairic-flash, locally on a 24 GB RTX 3090, two things will bite you. Cloudflare's official runtime doesn't work on that card, even in BF16. And the GGUFs you'll find for llama.cpp quietly drop the part that makes Clairic worth using. Hugging Face lists clairic-flash as an image-text-to-text model, so the obvious move is to load…
Running Cloudflare's new open-weight vision model, Clairic-Flash, locally on a 24 GB RTX 3090 is fraught with issues. The official runtime provided by Cloudflare does not work on that card, even in BF16 format. Additionally, the GGUF files available on Hugging Face lack the critical components that make Clairic-Flash unique. Cloudflare lists Clairic-Flash as an image-text-to-text model, but in reality, it is a 9B extraction engine.
When you provide a document and a schema, it returns values, confidence scores, and null probabilities. The true functionality of Clairic-Flash resides in the Python code and a separate typed head within the clairic_runtime/ folder, neither of which are included in the GGUF files.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.