TensorCore.si Blog
Tensor Cores, GPUs and AI chips, explained
In-depth guides, generation-by-generation breakdowns and analysis of the hardware that trains and runs modern AI.
Daily GPU Price TrackerH100 from $1.80/hr, H200 from $2.09/hr, B200 from $4.11/hr. Compare 31 cloud providers, updated daily..80/hr, H200 from .09/hr, B200 from $4.11/hr. Compare 31 cloud providers, updated daily.See prices →
Hopper vs Blackwell: What Changed for AI
H100 and H200 against B200 and B300: memory, bandwidth, FP8 and FP4 throughput, NVLink, and what the differences mean for training and for serving…
FP16, BF16, FP8 and FP4: Precision Formats Explained
Why AI runs on 16-, 8- and even 4-bit numbers, how each format trades range for precision, and what microscaling, NVFP4 and FP32 accumulation…
Tensor Cores vs CUDA Cores: What's the Difference?
CUDA cores do one multiply-add on single numbers; Tensor Cores do a whole matrix tile at once. Here is how they differ, how much…
How a GPU Works: SMs, Warps and the Memory Hierarchy
A tour of a modern NVIDIA GPU from the top down: streaming multiprocessors, warps of 32 threads, registers, shared memory, L2 cache and HBM,…