From c764aa0306a7cdfbf9bbf2e4ee077c3e02d9356f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?R=C3=A9mi=20Louf?= Date: Mon, 10 Jun 2024 14:49:14 +0200 Subject: [PATCH] Add statement regarding OS vs closed models --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index 4c3443639..f1bd7bc51 100644 --- a/README.md +++ b/README.md @@ -45,6 +45,7 @@ Outlines 〰 has new releases and features coming every week. Make sure to ⭐ s ## Why should I use structured generation? * It doesn't add any overhead during inference (cost-free) +* It allows Open Source models to beat closed source models ([Mistral](https://x.com/dottxtai/status/1797692104023363765), [GPT-4](https://x.com/dottxtai/status/1798443290913853770)) * [It speeds up inference](http://blog.dottxt.co/coalescence.html) * [It improves the performance of base models (GSM8K)](http://blog.dottxt.co/performance-gsm8k.html) * [It improves the performance of finetuned models (CoNNL)](https://predibase.com/blog/lorax-outlines-better-json-extraction-with-structured-generation-and-lora)