New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the Instinct MI440X run GPT-1?
Yes — here's how fast and up to what context, with real measured numbers.
GPT-1 on the Instinct MI440X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 6038k | 52769.2-64495.7 t/s | ~107.7-131.6 (slow) |
| Q8 | yes | 6040k | 105538.5-128991.5 t/s | ~215.4-263.2 (slow) |
| Q4 | yes | 6041k | 188461.5-230341.9 t/s | ~384.6-470.1 (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 GPT-1?▶
Yes. At Q4 it generates ~188461.5-230341.9 tok/s and fits up to 6041k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~384.6-470.1 tok/s.
How much context fits?▶
At Q4, up to about 6041k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.