Blackwell's 2-CTA MMA: one thread issues, two thread blocks compute
cfregly · x · 2026-11-11
cfregly explains B200 Blackwell's 2-CTA MMA: a single thread issues one 2-CTA matrix multiply-accumulate instruction (tcgen05.mma.ctagroup::2), and the two paired thread blocks (Cooperative Thread Arrays) share the math — one instruction, two CTAs doing the work.
More from Infra
- Apple reportedly building Harbor app to turn Macs into private AI inference servers with Thunderbolt 5 clustering — BenBajarin · 2026-11-11
- Agent Arena: Gemini 4 Carbon Gets More Cost-Competitive as Tasks Get Harder — rseroter · 2026-11-11
- AMD Zen 6 deep dive: cheaper branch mispredicts and a return to Zen 3/4 integer scheduling — jedisct1 · 2026-11-11
- Together AI SVP Dan Fu on kernels, serving 3T-param models day zero, and self-improving inference — togethercompute · 2026-11-11
- ExecuTorch adds WebGPU backend: run PyTorch models in-browser, up to 10x faster than CPU — PyTorch · 2026-11-11
- Hedra runs training experiments with AI agents steered by human researchers — testingcatalog · 2026-11-11