File size: 1,038 Bytes
06cf9c4
 
c5cafdd
06cf9c4
7a6e9de
 
06e607b
06cf9c4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
import gradio as gr
from llama_cpp import Llama
from huggingface_hub import hf_hub_download

a = hf_hub_download(repo_id="TheBloke/WizardLM-13B-V1.2-GGML", filename="wizardlm-13b-v1.2.ggmlv3.q5_1.bin")

llm = Llama((a,))

def generate_text(input_text):
    output = llm(f"Q: {input_text} A:", max_tokens=256, stop=["Q:", "\n"], echo=True)
    return output['choices'][0]['text']

input_text = gr.inputs.Textbox(lines= 10, label="Enter your input text")
output_text = gr.outputs.Textbox(label="Output text")

description = "llama.cpp implementation in python [https://github.com/abetlen/llama-cpp-python]"

examples = [
    ["What is the capital of France? ", "The capital of France is Paris."],
    ["Who wrote the novel 'Pride and Prejudice'?", "The novel 'Pride and Prejudice' was written by Jane Austen."],
    ["What is the square root of 64?", "The square root of 64 is 8."]
]

gr.Interface(fn=generate_text, inputs=input_text, outputs=output_text, title="Llama Language Model", description=description, examples=examples).launch()