Update app.py

app.py

@@ -8,11 +8,14 @@ import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 
 DESCRIPTION = """\
-# Llama 3.2 3B Instruct
+# EvaByte-SFT
 
-Llama 3.2 3B is Meta's latest iteration of open LLMs.
-This is a demo of [`meta-llama/Llama-3.2-3B-Instruct`](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct), fine-tuned for instruction following.
-For more details, please check [our post](https://huggingface.co/blog/llama32).
+EvaByte is a byte-level language model that combines multibyte prediction with the efficient EVA attention mechanism.
+This page hosts [`EvaByte/EvaByte-SFT`](https://huggingface.co/EvaByte/EvaByte-SFT), fine-tuned via supervised instruction data to enable chat and general instruction-following capabilities.  
+For full details on architecture, training recipe, and benchmarks, see their blog post and the project repository:
+
+- Blog: <https://hkunlp.github.io/blog/2025/evabyte>  
+- GitHub: <https://github.com/OpenEvaByte/evabyte>
 """
 
 MAX_MAX_NEW_TOKENS = 2048