DeepSparse Server
DeepSparse is an inference runtime taking advantage of sparsity with neural networks offering GPU-class performance on CPUs.
A YouTube video summarization app using Streamlit, Laingchain, Deepgram, and Mistral 7B
A Streamlit application to perform YouTube video summarization using Deepgram for superior audio transcription, Langchain for harvesting the power of the LLM, and Mistral 7B, a state-of-the-art and open-source LLM.
When deploying the YouTube Video Summarization app on Koyeb, the following environment variable will be provided. Take care to set the required variable with the appropriate value if not set.
Variable | Description | Default value | Required |
---|---|---|---|
DEEPGRAM_API_KEY | A Deepgram API key. | Yes |
Copy your Koyeb App public URL similar to https://<YOUR_APP_NAME>-<YOUR_KOYEB_ORG>.koyeb.app
from the Koyeb control panel and paste it into your browser.
You should land on the application web page where you can enter a YouTube video URL and launch the summarization.
DeepSparse is an inference runtime taking advantage of sparsity with neural networks offering GPU-class performance on CPUs.
Deploy Fooocus, a powerful AI image generation tool, on Koyeb
LangServe makes it easy to deploy LangChain applications as RESTful APIs.