Published May 18, 2021

Apache TVM and OctoML

Explore the cutting-edge advancements in AI with Luis Ceze, as he delves into specialized chips, adaptable architectures, and Apache TVM's role in optimizing AI models for diverse hardware platforms, while addressing performance and accuracy challenges in groundbreaking research and model deployment.

Episode Highlights

Topics covered

Episode Highlights

Related Episodes

MLOps is NOT Real
Answers 383 questions
Large models on CPUs
Answers 383 questions
End-to-end cloud compute for AI/ML
Answers 383 questions
MLOps and tracking experiments with Allegro AI
Answers 383 questions
Self-hosting & scaling models
Answers 383 questions
The fastest way to build ML-powered apps
Answers 383 questions
Automating code optimization with LLMs
Answers 383 questions
Generative models: exploration to deployment
Answers 383 questions
UBER and Intel’s Machine Learning platforms
Answers 383 questions
Operationalizing ML/AI with MemSQL
Answers 383 questions
Machine learning at small organizations
Answers 383 questions
Creating instruction tuned models
Answers 383 questions
TensorFlow in the cloud
Answers 383 questions
AI in the browser
Answers 383 questions
Mamba & Jamba
Answers 383 questions

Dexa/Practical AI

Apache TVM and OctoML

Topics covered

Popular Clips

Machine Learning Evolution

Machine Learning Innovations

AI Deployment Challenges

Simplifying Model Inference

Model Serialization Insights

Optimizing AI Discussions

Optimizing Machine Learning

Model Optimization Process

Community Engagement in AI

Future AI Innovations

Sparsity in AI Models

Getting Started with TVM

Journey into AI

Fun with AI

OctoML's Vision

Episode Highlights

Future AI Directions

Apache TVM Insights

Model Optimization Challenges

Related Episodes

MLOps is NOT Real

Large models on CPUs

End-to-end cloud compute for AI/ML

MLOps and tracking experiments with Allegro AI

Self-hosting & scaling models

The fastest way to build ML-powered apps

Automating code optimization with LLMs

Generative models: exploration to deployment

UBER and Intel’s Machine Learning platforms

Operationalizing ML/AI with MemSQL

Machine learning at small organizations

Creating instruction tuned models

TensorFlow in the cloud

AI in the browser

Mamba & Jamba

Apache TVM and OctoML

Topics covered

Popular Clips

Episode Highlights

Future AI DirectionsThe future of AI is being shaped by advancements in hardware and its integration with life sciences. Luis Ceze discusses the potential of specialized chips and adaptable architectures, as well as AI's role in enabling groundbreaking research in genomics and molecular design.

Future AI Directions

Apache TVM InsightsLuis Ceze, co-creator of Apache TVM, discusses how this innovative tool optimizes AI model performance across various hardware platforms. He highlights the role of community contributions and framework interoperability in enhancing the tool's capabilities.

Apache TVM Insights

Model Optimization ChallengesLuis Ceze explores the intricate balance between performance and accuracy in AI model optimization. He discusses the challenges of deploying models across various environments and the role of quantization in enhancing model efficiency.

Model Optimization Challenges

Related Episodes