🔌
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 A770 16GB run DEQ-Transformer (Post-LN) + Jacobian Regularisation?
Yes — here's how fast and up to what context, with real measured numbers.
DEQ-Transformer (Post-LN) + Jacobian Regularisation on the Intel Arc A770 16GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 199k | 1657.1-2057.1 t/s | ~118.4-146.9 (slow) |
| Q8 | yes | 200k | 3314.3-4114.3 t/s | ~236.7-293.9 (slow) |
| Q4 | yes | 201k | 5918.4-7346.9 t/s | ~422.7-524.8 (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.
→ All models on the Intel Arc A770 16GB
→ DEQ-Transformer (Post-LN) + Jacobian Regularisation specs
→ Other hardware
FAQ
Can the Intel Arc A770 16GB run DEQ-Transformer (Post-LN) + Jacobian Regularisation?▶
Yes. At Q4 it generates ~5918.4-7346.9 tok/s and fits up to 201k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~422.7-524.8 tok/s.
How much context fits?▶
At Q4, up to about 201k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.