🔌
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 MI430X run InternVL2_5-38B?
Yes — here's how fast and up to what context, with real measured numbers.
InternVL2_5-38B on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 1217k | 191.1-233.6 t/s | ~0.3-0.4 (slow) |
| Q8 | yes | 1364k | 382.3-467.2 t/s | ~0.7-0.8 (slow) |
| Q4 | yes | 1428k | 682.6-834.3 t/s | ~1.2-1.4 (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 MI430X run InternVL2_5-38B?▶
Yes. At Q4 it generates ~682.6-834.3 tok/s and fits up to 1428k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1.2-1.4 tok/s.
How much context fits?▶
At Q4, up to about 1428k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.