The Inference Bottleneck: Who Captures the Value?

Author: @fraaaankfu | IOSG Ventures Follow the GPUs: The $200B Question Becomes $600B In 2023, Sequoia's David Cahn named the question hanging over the entire AI build-out — the $200B question. For

Jun 9, 2026 · 9:49 AM UTC

1,550

Sort replies: Relevant Recent Liked

Kairav · Jun 10, 2026 · 5:12 PM UTC

Kairav @kairav_mis94620

Jun 10

Replying to @IOSGVC

We run classification, moderation, and extraction on quantized models at the edge via ZeroGPU instead of cloud inference APIs. Cuts per-call cost by 90% for workloads that don't need frontier reasoning.