Use-case guide · Long context

Should you pick GB300 NVL72 (per GPU) for long context?

GB300 NVL72 (per GPU) has 192 GB VRAM. Whether it's the right fit for long context depends on your model size, expected QPS, and budget. Below is what we're seeing in production.

VRAM + model fit

GB300 NVL72 (per GPU) fits models up to ~134B parameters in BF16 comfortably with room for KV-cache. For long context specifically, you'll want to leave headroom for context length growth.

Pricing

Live pricing across all providers for GB300 NVL72 (per GPU) is on the GPU detail page — click through for the sortable list.

Throughput

On long context workloads, GB300 NVL72 (per GPU) typically delivers the throughput published in its FP16 spec, minus the framework overhead (vLLM ≈ 85% MFU, TGI ≈ 70%).

Try the calculator to size the hardware for your specific model, or see all GPUs on the InferenceScore leaderboard.