Comment by Melatonic
19 hours ago
Does that include stuff like the Pro Blackwell 6000? Or are the tensor cores as good per SM comparably? They perform quite well on many tests
19 hours ago
Does that include stuff like the Pro Blackwell 6000? Or are the tensor cores as good per SM comparably? They perform quite well on many tests
Pro Blackwell 6000 is just a 5090 with more VRAM. It does not have the tcgen05 (5th gen tensor core) instructions despite the "5th gen tensor core) branding and thus do not support any optimized Blackwell (sm100) kernels.
Every Blackwell card other than the (G)B100, (G)B200, (G)B300 and Jetson Thor, use the Ampere tensor core instruction (mma.sync) but with fp4/6/8 added on. Beyond that the DGX Spark (which is advertised as having the same architecture as B200) has especially weak (not tcgen05) tensor cores that have a very narrow operating window and low utilization.