Latency: How long a single request takes. Measured in milliseconds. " latency is ms" means % of requests complete within ms.
Throughput: How many requests you handle per second. Measured in QPS or RPS.
They're related but different. You can have low latency but low throughput (fast but limited). Or high throughput but high latency (handles many requests, but each is slow).
Optimizing for one often hurts the other. Batching increases throughput but increases latency for individual requests.