🔌
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 Intel Arc B580 12GB run Tensor-Transformer(1core)+PN (WT103)?
Yes — here's how fast and up to what context, with real measured numbers.
Tensor-Transformer(1core)+PN (WT103) on the Intel Arc B580 12GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 143k | 1550.3-1924.5 t/s | ~136-168.8 (slow) |
| Q8 | yes | 144k | 3100.6-3849 t/s | ~272-337.6 (slow) |
| Q4 | yes | 145k | 5536.8-6873.2 t/s | ~485.7-602.9 (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 Intel Arc B580 12GB
→ Tensor-Transformer(1core)+PN (WT103) specs
→ Other hardware
FAQ
Can the Intel Arc B580 12GB run Tensor-Transformer(1core)+PN (WT103)?▶
Yes. At Q4 it generates ~5536.8-6873.2 tok/s and fits up to 145k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~485.7-602.9 tok/s.
How much context fits?▶
At Q4, up to about 145k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.