Faster than GPT and Free ,enjoy!

Performance of Zephyr 7B:

At the time of release, Zephyr-7B-β is the highest ranked 7B chat model on the MT-Bench and AlpacaEval benchmarks:

bench mark

In particular, on several categories of MT-Bench, Zephyr-7B-β has strong performance compared to larger open models like Llama2-Chat-70B:

Zephyr-7B-β

The three steps of our method:

(1) large scale, self-instruct-style dataset construction (UltraChat), followed by distilled supervised fine-tuning (dSFT), (2) AI Feedback (AIF) collection via an ensemble of chat model completions, followed by scoring by GPT-4 (UltraFeedback) and binarization into preferences, and (3) distilled direct preference optimization (dPO) of the dSFT model utilizing the feedback data.

Technique used: