New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the NVIDIA GeForce RTX 3080 10GB run MegaMolBART?
Yes — here's how fast and up to what context, with real measured numbers.
MegaMolBART on the NVIDIA GeForce RTX 3080 10GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 116k | 6164.4-7346.7 t/s | ~324.4-386.7 (slow) |
| Q8 | yes | 117k | 12328.9-14693.3 t/s | ~648.9-773.3 (slow) |
| Q4 | yes | 117k | 22015.9-26238.1 t/s | ~1158.7-1381 (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 NVIDIA GeForce RTX 3080 10GB run MegaMolBART?▶
Yes. At Q4 it generates ~22015.9-26238.1 tok/s and fits up to 117k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1158.7-1381 tok/s.
How much context fits?▶
At Q4, up to about 117k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.