GPU × use-case guide · Text classification

Is the H100 SXM a good GPU for text classification?

The H100 SXM is a hopper NVIDIA GPU with 80 GB HBM3 (3350 GB/s), 990 TFLOPS BF16 / 1979 TFLOPS FP8, and a 700 W TDP. Text classification workloads care most about very high request throughput on small models — this is a cost-per-1k-requests game, not a VRAM game. Here's how the H100 SXM measures up.

What models fit on a single H100 SXM?

Weights only, reserving ~25% of the 80 GB for KV cache, activations and fragmentation. ✓ = fits on one card.

Largest single-card fit: Gemma 2 27B at BF16, Mixtral 8x7B (MoE) at FP8, Qwen 2.5 72B at INT4. Bigger models need tensor-parallel across 8 cards.

H100 SXM for text classification, specifically

Text classification is throughput-bound rather than VRAM-bound, so the H100 SXM's 3350 GB/s of bandwidth and 990 TFLOPS matter more than raw capacity. Because it runs offline, batch aggressively to push tokens-per-dollar down. Size it precisely on the calculator.

H100 SXM pricing across providers

Verdict

With 80 GB, the H100 SXM is a data-center-class card that comfortably handles text classification for models up to Gemma 2 27B at full precision on a single card — a strong pick if your budget supports ~$2.49/hr.