PANEL - Data, Code, and Life with LLMs.
summary
One-hour debate argues whether new billion‑dollar startups in five years come from open or closed models, citing concrete evidence: Llama 2’s released weights plus llama.cpp speedups let teams run models on far fewer GPUs (claimed down to one or two A100s), GPT‑4 remains the fastest path to market for many builders, Databricks and RAG tooling support production use, and services like Pulse AI aggregate multiple model APIs. Key tensions are cost and latency tradeoffs, fine‑tuned proprietary data that can’t be shared, and who controls massive compute investments.
AI-written summary · about
chapters (9)
- 0:00Panel introduction
- 1:13Debate proposition explained
- 3:39Open source team introductions
- 11:13Closed source team introductions
- 19:20Open source team rebuttal
- 30:46Closed source team rebuttal
- 42:03Audience Q&A begins
- 50:26Final arguments and closing statements
- 58:35Final vote and conference wrap-up
related talks
12mKeynote: How LLMs Have Changed the Way Organizations Do AI
42mOpen Source LLMs — Powering AI Accessibility and Innovation
12mKeynote: The Llama Ecosystem: Past, Present and Future
24mIndustry Panel- 12mKeynote: How LLMs Have Changed the Way Organizations Do AI
22mThe Disruptive Potential of On-Device LLMs
36mInsights from The State of AI Report 2023
33mOpening Keynote: LLMs, between fantasy and reality - the poin...