New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the NVIDIA H200 run BLOOM-560M?
Yes — here's how fast and up to what context, with real measured numbers.
BLOOM-560M on the NVIDIA H200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1939k | 3128.6-3728.6 t/s | ~26.1-31.1 (slow) |
| Q8 | yes | 1948k | 6257.1-7457.1 t/s | ~52.1-62.1 (slow) |
| Q4 | yes | 1952k | 11173.5-13316.3 t/s | ~93.1-111 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to system RAM offload (depends on your system RAM) and speed drops to the "beyond" figure.
FAQ
Can the NVIDIA H200 run BLOOM-560M?▶
Yes. At Q4 it generates ~11173.5-13316.3 tok/s and fits up to 1952k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~93.1-111 tok/s.
How much context fits?▶
At Q4, up to about 1952k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.