🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN — FITS
Can the NVIDIA B200 run StableLM-3B-4E1T?
Yes — here's how fast and up to what context, with real measured numbers.
StableLM-3B-4E1T on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2587k | 1044.6-1244.9 t/s | ~5.2-6.2 (slow) |
| Q8 | yes | 2630k | 2089.1-2489.8 t/s | ~10.4-12.4 (slow) |
| Q4 | yes | 2649k | 3730.6-4446 t/s | ~18.7-22.2 (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 StableLM-3B-4E1T?▶
Yes. At Q4 it generates ~3730.6-4446 tok/s and fits up to 2649k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~18.7-22.2 tok/s.
How much context fits?▶
At Q4, up to about 2649k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.