🔌
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 Instinct MI440X run GLIDE?
Yes — here's how fast and up to what context, with real measured numbers.
GLIDE on the Instinct MI440X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 5935k | 1764-2156 t/s | ~3.6-4.4 (slow) |
| Q8 | yes | 5988k | 3528-4312 t/s | ~7.2-8.8 (slow) |
| Q4 | yes | 6012k | 6300-7700 t/s | ~12.9-15.7 (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 Instinct MI440X run GLIDE?▶
Yes. At Q4 it generates ~6300-7700 tok/s and fits up to 6012k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~12.9-15.7 tok/s.
How much context fits?▶
At Q4, up to about 6012k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.