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 Tensor-Transformer(1core)+PN (WT103)?
Yes — here's how fast and up to what context, with real measured numbers.
Tensor-Transformer(1core)+PN (WT103) on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2670k | 34232.1-40797.2 t/s | ~171.2-204 (slow) |
| Q8 | yes | 2671k | 68464.2-81594.4 t/s | ~342.3-408 (slow) |
| Q4 | yes | 2672k | 122257.6-145704.2 t/s | ~611.3-728.5 (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 Tensor-Transformer(1core)+PN (WT103)?▶
Yes. At Q4 it generates ~122257.6-145704.2 tok/s and fits up to 2672k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~611.3-728.5 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.