New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the Instinct MI455X run DeepSeek-R1-Distill-Qwen-1.5B?
Yes — here's how fast and up to what context, with real measured numbers.
DeepSeek-R1-Distill-Qwen-1.5B on the Instinct MI455X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 5987k | 3468.5-4239.3 t/s | ~7.1-8.7 (slow) |
| Q8 | yes | 6014k | 6937.1-8478.7 t/s | ~14.2-17.3 (slow) |
| Q4 | yes | 6026k | 12387.6-15140.4 t/s | ~25.3-30.9 (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 DeepSeek-R1-Distill-Qwen-1.5B?▶
Yes. At Q4 it generates ~12387.6-15140.4 tok/s and fits up to 6026k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~25.3-30.9 tok/s.
How much context fits?▶
At Q4, up to about 6026k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.