🔌
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 Huawei Ascend 950DT run NVIDIA Nemotron 3 Nano Omni?
Yes — here's how fast and up to what context, with real measured numbers.
NVIDIA Nemotron 3 Nano Omni on the Huawei Ascend 950DT
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 259k | 41.1-50 t/s | ~0.4-0.5 (slow) |
| Q8 | yes | 379k | 82.3-100 t/s | ~0.8-1 (slow) |
| Q4 | yes | 432k | 146.9-178.6 t/s | ~1.5-1.8 (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 Huawei Ascend 950DT run NVIDIA Nemotron 3 Nano Omni?▶
Yes. At Q4 it generates ~146.9-178.6 tok/s and fits up to 432k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1.5-1.8 tok/s.
How much context fits?▶
At Q4, up to about 432k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.