Dylan Patel on

cost reduction

2 entries, 27 Mar 2025 to 16 Jul 2026

On the recordsourced and dated, oldest first

    1. spoken

      Patel states same-quality AI model output became 1200x cheaper than GPT-3 through successive model releases.

      “Anthropic released new models. Google released new models. Meta released new models. Right? And then now it is 1200x times it's 1200x cheaper for the same output.”

      27 Mar 2025 · MedBricks Webcast · 31:05 · source · permalink
    1. spoken

      Patel argues layering KV cache offload and multi-token prediction on open-source engines drastically cuts inference costs.

      “So if you take the, you know, just open source inference engines off the shelf, that gets you a certain level of cost.”

      16 Jul 2026 · RAISE Summit · 14:24 · source · permalink

Dylan Patel ontheir other subjects

Everything Dylan Patel is on record saying · RSS