StartupXO
Search
English

Find XO: Cloudflare Releases Clef, an Open Source Decision Model Supporting Fast and Accurate Classification, and a Reinforcement Learning Fine-Tuning Platform (blog.cloudflare.com)

No votes yet startupxo 1 comment

Writing language: Korean Read in the original language

Summary / Read source ↗

- Clef is a decision model that accepts text and image inputs, selects one category among predefined options, and provides probabilities for each choice, helping automate complex decision-making tasks. - In Cloudflare's internal evaluations, Clef and Clef-flash models demonstrated equal or superior accuracy compared to the competing Jev model across multiple quality benchmarks, with processing latencies of an average of 209ms for Clef and 39ms for Clef-flash (according to Cloudflare evaluation results). - The Clef model is deployed in the Cloudflare Workers AI environment to reduce network latency and offers full API compatibility with the existing Jev model, making it easy to adopt in existing systems. - Cloudflare announced a reinforcement learning-based fine-tuning service for task-specific customization of Clef and is preparing a platform that will allow customers to manage everything from data collection to fine-tuning and redeployment themselves in the future.
Found on

Hacker News ↗ / 442 votes / 163 comments

Sign in to comment

1 comment

Editorial opinion startupxo

If it's a decision model that supports image input from the start and can handle 64k tokens, unlike general text-based classification, I'd like to experience how effectively it can handle complex tasks utilizing visual information. Especially since it runs on Cloudflare's edge infrastructure with short response times, that's a notable point to experience firsthand. However, fine-tuning is currently only available through partner collaboration, so if you want to try custom adjustments directly, you'll have to wait a bit longer.
Writing language: Korean

Keyboard shortcuts

Choose a post with the up and down arrows, then press Enter.

↑ / ↓
Previous post / next post
Enter
Open summary and comments for the selected post
Tab
Move to the submit or comment button, then press Enter

Type normally in text fields. Tab and Enter are always available.