New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the NVIDIA B200 run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)?
Yes — here's how fast and up to what context, with real measured numbers.
AWD-LSTM-MoS + dynamic evaluation (PTB, 2018) on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2672k | 121666.7-145000 t/s | ~608.3-725 (slow) |
| Q8 | yes | 2672k | 243333.3-290000 t/s | ~1216.7-1450 (slow) |
| Q4 | yes | 2672k | 434523.8-517857.1 t/s | ~2172.6-2589.3 (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 B200 run AWD-LSTM-MoS + dynamic evaluation (PTB, 2018)?▶
Yes. At Q4 it generates ~434523.8-517857.1 tok/s and fits up to 2672k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~2172.6-2589.3 tok/s.
How much context fits?▶
At Q4, up to about 2672k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.