AMD says Instinct MI355X cut p99 latency in a Signal_65 RAG test
AMD says the Instinct MI355X delivered about 42% lower p99 latency and roughly twice the concurrent-user headroom before breaching the evaluated SLA in Signal_65 RAG testing hosted via TensorWave. The available statement does not name the comparison system or the SLA threshold.
AMD points to the Signal_65 RAG test hosted via TensorWave for the result.