GPU × use-case guide · Summarisation

Is the H100 SXM a good GPU for summarisation?

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. Summarisation workloads care most about large context windows and batch throughput — documents are long and jobs run offline, so cost-per-token beats latency. 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 summarisation, specifically

Summarisation is context-heavy, so the KV cache — not the weights — is what fills the 80 GB. On the H100 SXM you'll trade context length against batch size: long prompts mean fewer concurrent requests. 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 summarisation for models up to Gemma 2 27B at full precision on a single card — a strong pick if your budget supports ~$2.49/hr.