B200 SXM 180GB vs B300 SXM 288GB for LLM inference

B200 SXM 180GB vs B300 SXM 288GB: verdict

B200 SXM 180GB vs B300 SXM 288GB: the short version is that B300 SXM 288GB wins on VRAM (180GB on the B200 SXM 180GB vs 288GB on the B300 SXM 288GB), B300 SXM 288GB wins on raw memory bandwidth (7700 GB/s on the B200 SXM 180GB vs 8000 GB/s on the B300 SXM 288GB, the number that dominates decode throughput), and B200 SXM 180GB is the cheaper of the two to rent today. Which of those matters more depends entirely on your workload: a large model or long-context deployment cares most about the VRAM number, while a smaller model serving many short requests cares much more about bandwidth and cost per token. When VRAM is tied, bandwidth and price become the primary levers. Neither the B200 SXM 180GB nor the B300 SXM 288GB is a universally correct answer — they're built for overlapping but distinct points on the capacity-versus-throughput curve, and the gap between them (in bandwidth and price, and potentially VRAM) is usually the whole reason to pick one over the other rather than defaulting to whichever one you already have access to. The full spec table above covers bf16/fp8 compute throughput, NVLink support, and power draw for both cards, and the models table above shows which currently-popular open-weight models actually fit on each one — the more concrete way to answer 'which GPU do I need' than specs alone.