🔌
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 Qwen2.5-Coder (7B)?
Yes — here's how fast and up to what context, with real measured numbers.
Qwen2.5-Coder (7B) on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2905k | 964.5-1178.8 t/s | ~1.7-2 (slow) |
| Q8 | yes | 2963k | 1928.9-2357.6 t/s | ~3.3-4 (slow) |
| Q4 | yes | 2988k | 3444.5-4209.9 t/s | ~5.9-7.2 (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 MI430X run Qwen2.5-Coder (7B)?▶
Yes. At Q4 it generates ~3444.5-4209.9 tok/s and fits up to 2988k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~5.9-7.2 tok/s.
How much context fits?▶
At Q4, up to about 2988k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.