🔌
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 Transformer (Adaptive Input Embeddings) WT103?
Yes — here's how fast and up to what context, with real measured numbers.
Transformer (Adaptive Input Embeddings) WT103 on the Huawei Ascend 950DT
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1991k | 5263.2-6396.8 t/s | ~52.6-64 (slow) |
| Q8 | yes | 1995k | 10526.3-12793.5 t/s | ~105.3-127.9 (slow) |
| Q4 | yes | 1996k | 18797-22845.6 t/s | ~188-228.5 (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.
→ All models on the Huawei Ascend 950DT
→ Transformer (Adaptive Input Embeddings) WT103 specs
→ Other hardware
FAQ
Can the Huawei Ascend 950DT run Transformer (Adaptive Input Embeddings) WT103?▶
Yes. At Q4 it generates ~18797-22845.6 tok/s and fits up to 1996k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~188-228.5 tok/s.
How much context fits?▶
At Q4, up to about 1996k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.