Costs of AI models

3 videos across 3 channels

Viewers dive into the practical price tag of running AI models. One piece tests DeepSeek V4 Pro with Claude Code Router, weighing its costs against Kimmy, GLM, and Miniax, and breaking down plan modes and open-weight options. Another video dissects Mythos—its benchmarks, security concerns, and marketing hype—while a Laravel AI SDK session maps out costs for building a document QA bot with vector stores and embeddings, and teases local PostgreSQL vector search as a future topic.

DeepSeek V4 Pro Inside Claude Code = GOD MODE (FREE?) thumbnail

DeepSeek V4 Pro Inside Claude Code = GOD MODE (FREE?)

The video documents a test run using Claude Code to integrate DeepSeek (V4 Pro) and explores how to set it up with Claud

00:08:22
🔴 Claude Mythos Preview thumbnail

🔴 Claude Mythos Preview

The streamer breaks down Mythos and its hype, reviews the latest news and benchmarks, and weighs security concerns and m

01:24:03
Laravel AI SDK: Q&A with Internal Docs (Stores / File Search / Embeddings) thumbnail

Laravel AI SDK: Q&A with Internal Docs (Stores / File Search / Embeddings)

This video demonstrates how to use the Laravel AI SDK to build a chatbot that answers questions from an uploaded documen

00:08:21