NextArete: Global Tech Frontier

Topic

Inference

A collection of 3 issues

The Chip War in the Inference Era: From Peak Specs to Tokens per Watt per Dollar, and What Blackwell Ultra vs MI355X Signals

NVIDIA cites SemiAnalysis InferenceX to frame Blackwell Ultra around efficiency and cost, while AMD highlights MI355X throughput in interactive ranges with latency-reducing techniques. The real battle is inference economics and full-stack execution.

NVIDIA Cites SemiAnalysis InferenceX: Blackwell Ultra Shifts the Battle to Tokens per Watt and Cost per Token for Agentic AI Inference

NVIDIA points to SemiAnalysis InferenceX data to argue Blackwell Ultra improves efficiency for agentic inference, reframing competition around tokens per watt and cost per token.

AMD on MI355X Inference: Multi-Token Prediction to Reduce Latency, Focus on Throughput in Interactive Ranges

AMD’s technical article outlines MI355X inference optimizations such as multi-token prediction, framing performance around throughput under interactive latency constraints.