🔌
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 Sailor-7B-Chat?
Yes — here's how fast and up to what context, with real measured numbers.
Sailor-7B-Chat on the NVIDIA B200
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1218k | 378.2-450.8 t/s | ~1.9-2.3 (slow) |
| Q8 | yes | 1277k | 756.5-901.6 t/s | ~3.8-4.5 (slow) |
| Q4 | yes | 1303k | 1350.9-1609.9 t/s | ~6.8-8 (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 Sailor-7B-Chat?▶
Yes. At Q4 it generates ~1350.9-1609.9 tok/s and fits up to 1303k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~6.8-8 tok/s.
How much context fits?▶
At Q4, up to about 1303k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.