TruEra Challenge: Build LLM Applications

Let's Kick it Off! We are starting the stream with a Hackathon opening, an introduction to lablab.ai from the NewNative co-founder, and a schedule overview.

Hackathon Kick-Off Stream

Join Anupam Datta, Co-founder, President, and Chief Scientist at TruEra, as he kicks off the Hackathon with insightful opening words, setting the tone for an engaging exploration into TruEra Challenge: Build LLM Applications with Google Vertex AI.

TruEra Opening Words

Let's dive into the platform and its features that will be useful for Hackathon participants.

Hackathon Guide and Platform Walk-through

Join the workshop with Michael Davis, Customer Engineer at Google and Anupam Datta, co-founder, President, and Chief Scientist at TruEra.

Google Cloud x TruEra Workshop: Running LLM-based Evals with PaLM 2

Join us for an insightful workshop featuring Josh Reini, DevRel Engineer at TruEra, and Yujian Tang, Developer Advocate at Zilliz.

Build and Iterate with Advanced Index Construction: Chunking, Range Search and more

General Q&A session on Discord

In this workshop on December 4, Shayak Sen, CTO of Truera, and Laurie Voss, VP of Developer Relations at LlamaIndex, will explore the building and evaluation of multimodal GenAI apps using the RAG Triad model. 

Building and Evaluating Multimodal GenAI Apps: RAG Triad is All you Need

Submissions Deadline

Extended Submission Deadline💻

TOP-5 teams will be invited to pitch their ideas for Hackathon judges - to define the Hackathon winner!

Finalists Stream

🌟Team Tonic Uses Vertex to serve + fine tune T5 Models and TruEra to optimize multimodal inputs💪🏻

Team Tonic

Tonic

Hackathon Submission: Enhanced Multimodal AI Performance

Project Title: Optimizing Multimodal AI for Real-World Applications

Overview:
Our project focused on optimizing multimodal AI performance using the TruEra Machine Learning Ops platform. We evaluated 18 models across vision, audio, and text domains, employing innovative prompting strategies, performance metrics, and sequential configurations.

Methodology:

Prompting Strategies: Tailored prompts to maximize model response accuracy.
Performance Metrics: Assessed models on accuracy, speed, and error rate.
Sequential Configurations: Tested various model combinations for task-specific effectiveness.
Key Models Evaluated:

Vision: GPT4V, LLava-1.5, Qwen-VL, Clip (Google/Vertex), Fuyu-8B.
Audio: Seamless 1.0 & 2.0, Qwen Audio, Whisper2 & Whisper3, Seamless on device, GoogleAUDIOMODEL.
Text: StableMed, MistralMed, Qwen On Device, GPT, Mistral Endpoint, Intel Neural Chat, BERT (Google/Vertex).
Results:

Top Performers: Qwen-VL in vision, Seamless 2.0 in audio, and MistralMed in text.
Insights: Balance between performance and cost is crucial. Some models like GPT and Intel Neural Chat underperformed or were cost-prohibitive.
Future Directions:

Focus on fine-tuning models like BERT using Vertex.
Develop more connectors for TruLens for diverse endpoints.
Submission Contents:

GitHub Repository: [Link]
Demo: [Link]
Presentation: [Link]
Our submission showcases the potential of multimodal AI evaluation using TruEra / TruLens in enhancing real-world application performance, marking a step forward in human-centered AI solutions.

From the award winning makers of TruEra Enhanced , 

🔬TruEra🧫Applied

This submission is more about the journey than the destination. We took an existing application and asked some tough questions with painless answers from Trulens & made it better.

"Team Tonic team on TruEra Challenge: Build LLM Applications Hackathon"

Team Idea

Submission