#open-source-ai
17 episodes
#3020: How Chatterbox Locks Your Voice Clone Across Thousands of Generations
Why most single-shot TTS models drift over time—and how Chatterbox's cached embedding approach solves it.
#2426: Why DeepSeek V4's Prose Feels More Vivid Than Claude or GPT
A million-token context window at 2% the KV-cache cost — and prose that actually breathes. Here's what makes V4 different.
#2016: Andrej Karpathy: The Bob Ross of Deep Learning
Why the most influential AI mind prefers a blank text file to proprietary black boxes.
#2009: The Plumbing of AI Safety: Guardrails, Not Vibes
We dive deep into the specific libraries, proxy layers, and architectural decisions that keep an LLM from emptying a bank account.
#1947: Curation Is the New Creation
With 47 new AI video tools launching in a week, finding the right one is harder than using it.
#1940: Why Google's 31B Model Fits in Your GPU
Google just dropped Gemma four, and its 31-billion-parameter size is a masterclass in hardware-aware AI design.
#1919: Android Dev Without Android Studio: Is It Actually Good?
How to ship an Android app without ever opening Android Studio or touching a line of Java.
#1808: The Architecture That Made AI Voices Run on a Raspberry Pi
How a model the size of a tweet outperforms billion-dollar giants in the race for perfect AI speech.
#1737: Nous Research: The Decentralized AI Lab Beating Giants
Meet Nous Research, the decentralized collective outperforming billion-dollar labs with open-source AI and the self-improving Hermes-Agent framework.
#1736: The Hidden AI Economy: Following the Tokens
OpenClaw is processing 16.5 trillion tokens daily, dwarfing Wikipedia. Here’s why it’s #1.
#1711: Stop Building the Bucket: The Vendor SDK Era
We compare the three major vendor SDKs for building AI agents, weighing speed, safety, and scalability.
#1668: Kimi K2's Hidden Reasoning: A New AI Architecture
Moonshot AI's Kimi K2 Thinking model uses a hidden reasoning phase to solve complex logic puzzles and coding tasks, beating top proprietary models.
#1632: Agent Interview: DeepSeek V three point two
We interview DeepSeek V3 to see if this open-weight powerhouse can handle weird podcast prompts better than big tech’s flagship models.
#1561: Abliteration: The High-Dimensional Lobotomy of AI
Discover how researchers are surgically removing refusal filters from AI models using a mathematical process called abliteration.
#847: Abliterating the AI Schoolmarm: Who Owns Your LLM?
Explore why users are ditching corporate AI for "uncensored" local models and how "refusal vectors" are being mathematically removed.
#107: The $5.5 Million Breakthrough: DeepSeek’s AI Disruption
Discover how DeepSeek-V3 is disrupting the AI market with massive cost savings and technical innovations like Multi-Head Latent Attention.
#87: Why Big Tech Gives Away AI for Free
Why are tech giants spending millions on AI just to give it away? Herman and Corn dive into the strategic chess game of open-source models.