Skip to content
← Inference System Design / 15
hardNewOpenAIDatabricksSnowflake

Design batch inference over 1 billion documents: embed each one and run a 70B model over each. How many GPUs, how long, what cost?

Offline work changes the regime: the batch is as big as memory allows, latency is irrelevant, and output tokens stop being expensive. The answer prices both passes in GPU-hours, finds which one is the bill, and cuts it.

Updated Oct 2026 · Worked answers on inference mechanisms, with calculations, implementation details, and follow-up questions.

Offline work changes the regime: the batch is as big as memory allows, latency is irrelevant, and output tokens stop being expensive. The answer prices both passes in GPU-hours, finds which one is the bill, and cuts it.

10 answers per topic instead of 5, plus saved progress and bookmarks · costs nothingor unlock all 136 remaining answers · ₹2,000 / $25

Every worked answer and derivation in all 12 topics, every concept page, and the full practice-test bank.

6 months · One payment · No auto-renewal

UP NEXT ON YOUR JOURNEY
Next in this trackCut inference cost per million output tokens by 70% while preserving quality and latency. Can you defend the target?Popular right nowWhy is batch-1 decode on a 70B model bandwidth-bound, and what is its latency floor on two H100s?Popular right nowDerive the FLOPs per token for a dense transformer from its config. Where does the 2N rule stop being accurate?
DISCUSSION · 0

Nobody has commented on this answer. Add a correction or another way to work it.