From 9d3ee016c938738cdba705e3b37be56ceee0887b Mon Sep 17 00:00:00 2001
From: Juarez Bochi <juarez.bochi@grammarly.com>
Date: Mon, 18 Dec 2023 08:50:36 -0500
Subject: [PATCH] Add readme.md for t5

---
 t5/README.md | 29 +++++++++++++++++++++++++++++
 1 file changed, 29 insertions(+)
 create mode 100644 t5/README.md

diff --git a/t5/README.md b/t5/README.md
new file mode 100644
index 00000000..5b173d42
--- /dev/null
+++ b/t5/README.md
@@ -0,0 +1,29 @@
+# T5
+
+[T5](https://arxiv.org/pdf/1910.10683.pdf) are encoder-decoder models pre-trained on a multi-task mixture of unsupervised and supervised tasks. T5 works well on a variety of tasks out-of-the-box by prepending a different prefix to the input corresponding to each task, e.g.: `translate English to German: …`, `summarize: ….`
+
+## Setup
+
+Download and convert the model:
+
+```sh
+python convert.py --model t5-small
+```
+
+This will make the `{model}.npz` file which MLX can read.
+
+## Generate
+
+To run the model, use the `t5.py` script:
+
+```sh
+python t5.py --model t5-small --prompt "translate English to German: A tasty apple"
+```
+
+Should give the output: `Ein schmackhafter Apfel`
+
+To see a list of options run:
+
+```sh
+python t5.py --help
+```