Moreh says AMD MI355X delivered over twice Nvidia B200's inference throughput in tests

South Korean AI software developer Moreh reports that AMD Instinct MI355X GPUs running its inference software achieved more than twice the throughput of Nvidia B200 GPUs in company tests.

Read the full story at DigiTimes Asia ↗

Timeline · 1 report

2026-11-05 03:28 · DigiTimes Asia
South Korea's Moreh claims AMD MI355X delivers over twice Nvidia B200's inference throughput

More stories

AMD forecasts quarterly revenue above estimates on AI data-center demand — Reuters AI
Samsung to make AMD and Nvidia HBM parts at second Texas fab, report says — The Next Web
Anthropic secures over 3.5GW of compute capacity for 2026, adding Nvidia and AMD to procurement — Techmeme
AMD takes $1.07 billion quarterly charge tied to OpenAI and Meta chip deals — Reuters AI
Nvidia's DSX POD gains traction as AI data-center buildout accelerates — Barron's
AI inference startup Fal raises $1.3B at $16B valuation in Bessemer-led round — Bloomberg AI
NVIDIA Details AI Agent Blueprints for Factory and AI Data Center Operations — NVIDIA Technical Blog
Revolut says it cut AI coding inference costs by about 90% by self-hosting GLM models — The Register AI/ML

Get the daily brief of stories like this at 6:30 every morning →