🔌
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 MI455X run Incoder-6.7B?
Yes — here's how fast and up to what context, with real measured numbers.
Incoder-6.7B on the Instinct MI455X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2919k | 921.5-1126.3 t/s | ~1.9-2.3 (slow) |
| Q8 | yes | 2970k | 1843-2252.5 t/s | ~3.8-4.6 (slow) |
| Q4 | yes | 2992k | 3291-4022.4 t/s | ~6.7-8.2 (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 MI455X run Incoder-6.7B?▶
Yes. At Q4 it generates ~3291-4022.4 tok/s and fits up to 2992k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~6.7-8.2 tok/s.
How much context fits?▶
At Q4, up to about 2992k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.