The AI Infrastructure Stack
Overview  /  Tier II Compute as a Service  /  Layer 04: AI Cloud & Inference
Sub-category 4.3

Inference Specialists

Custom-silicon or highly-optimized inference clouds competing on speed and cost-per-token.

Players

Players: Groq Private, Cerebras Private (S-1 filed), SambaNova Private, Together AI Private, Fireworks AI Private, Baseten Private, Modal Private, RunPod Private, Replicate Private, Deep Infra Private

Analysis coming soon — this page is scaffolding for deeper research into inference specialists.