Gradio-llama2.mojo / gradio_app.py
radames's picture
new
c7eecb3
raw
history blame
908 Bytes
import gradio as gr
import subprocess
import sys
import os
async def generate(prompt):
# os.environ["PROMPT"] = prompt
# stream stout
process = subprocess.Popen(
["mojo", "llama2.mojo"], stdout=subprocess.PIPE, stderr=subprocess.PIPE
)
text = ""
for char in iter(lambda: process.stdout.read(1), b""):
char_decoded = char.decode()
sys.stdout.write(char_decoded)
text += char_decoded
yield text
output_text = gr.Textbox(label="Generated Text")
demo = gr.Interface(
fn=generate,
inputs=None,
outputs=output_text,
description="""
# llama2.🔥
## [Mojo](https://docs.modular.com/mojo/) implementation of [llama2.c](https://github.com/karpathy/llama2.c) by [@tairov](https://github.com/tairov)
Source: https://github.com/tairov/llama2.mojo
""",
allow_flagging="never",
)
demo.queue()
demo.launch(server_name="0.0.0.0")