🔌
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 Qwen3-0.6B?
Yes — here's how fast and up to what context, with real measured numbers.
Qwen3-0.6B on the Instinct MI440X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 6023k | 10290-12576.7 t/s | ~21-25.7 (slow) |
| Q8 | yes | 6032k | 20580-25153.3 t/s | ~42-51.3 (slow) |
| Q4 | yes | 6036k | 36750-44916.7 t/s | ~75-91.7 (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 Qwen3-0.6B?▶
Yes. At Q4 it generates ~36750-44916.7 tok/s and fits up to 6036k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~75-91.7 tok/s.
How much context fits?▶
At Q4, up to about 6036k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.