🔌
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 Apple M3 Max 36GB 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 Apple M3 Max 36GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 469k | 1195.8-1442 t/s | ~39.9-48.1 (slow) |
| Q8 | yes | 470k | 2391.6-2883.9 t/s | ~79.7-96.1 (slow) |
| Q4 | yes | 471k | 4270.6-5149.9 t/s | ~142.4-171.7 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.
FAQ
Can the Apple M3 Max 36GB run Tensor-Transformer(1core)+PN (WT103)?▶
Yes. At Q4 it generates ~4270.6-5149.9 tok/s and fits up to 471k context; beyond that it spills to SSD swap and slows to ~142.4-171.7 tok/s.
How much context fits?▶
At Q4, up to about 471k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.