Discussing LLM Voice Workflows: Long Rambles Work, But Cost Remains a Barrier

techartist_ · x · 2026-07-22

Replying to Andrej Karpathy, a user agreed that Karpathy's suggested LLM workflow—switching to voice mode for a 10-minute, stream-of-consciousness ramble to feed the model enough context—matches their exact sentiment.

However, they also pointed out that the rising costs of using LLMs make this highly token-consuming scenario less accessible for most common folk.

Related event: Karpathy Recommends 10-Minute Voice Rambling for LLM Context(17 posts)→

Original post →

More from Apps

Apps channel →