metadata
title: README
emoji: 🔥
colorFrom: gray
colorTo: blue
sdk: static
pinned: false
SambaNova & Gradio Llama-thon!
SambaNova is a hardware company and has developed an extremely efficient chip, the SN40L, which is optimized to deliver extremely fast inference on AI models big and small. They recently released SambaNova Cloud, a cloud platform delivering extremely fast inference on the latest Llama open-source models, including Llama 3.1 405B – 10X faster than GPUs.
The Challenge:
SambaNova has partnered with Gradio to develop a custom integration to make it very fast and easy for developers to deploy Gradio applications to Hugging Face Spaces called Sambanova Gradio. Speed and efficiencies are a key part of this collaboration. Given these values, developers should build the best Hugging Face space deployed in the Hackathon organization that can improve their productivity with the speed and efficiency of SambaNova.Some Idea:
• Using the accuracy and speed of 405B, build a coding assistant that helps you iteratively code an application • Using the Llama 3.2 vision capabilities, create an application that analyzes an image and builds a front end interface • Using other models deployed on Hugging Face, create an end to end meeting note taker / assistantExamples of Spaces using sambanova and gradio:
• Llama Code Editor• AnyChat
Prizes:
Prizes will be given to top 3 spaces by number of likes:1st place:
• $5000 in Huggingface credits• SambaNova Free Developer Tier for a month
• Social promotion
2nd place:
• $3000 in Huggingface credits• SambaNova Free Developer Tier for a month
• Social promotion
3rd place:
• $2000 in Huggingface credits• SambaNova Free Developer Rate Limits for a month
• Social Promotion