New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the NVIDIA L40S 48GB run Novae?
Yes — here's how fast and up to what context, with real measured numbers.
Novae on the NVIDIA L40S 48GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 650k | 9855-11745 t/s | ~456.3-543.8 (slow) |
| Q8 | yes | 650k | 19710-23490 t/s | ~912.5-1087.5 (slow) |
| Q4 | yes | 651k | 35196.4-41946.4 t/s | ~1629.5-1942 (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 L40S 48GB run Novae?▶
Yes. At Q4 it generates ~35196.4-41946.4 tok/s and fits up to 651k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1629.5-1942 tok/s.
How much context fits?▶
At Q4, up to about 651k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.