🔌
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 NVIDIA GeForce RTX 5090 run FragLlama: Next-fragment prediction for molecular design?
Yes — here's how fast and up to what context, with real measured numbers.
FragLlama: Next-fragment prediction for molecular design on the NVIDIA GeForce RTX 5090
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 106k | 93.4-111.4 t/s | ~2.1-2.5 (slow) |
| Q8 | yes | 160k | 186.9-222.7 t/s | ~4.2-5 (slow) |
| Q4 | yes | 183k | 333.7-397.7 t/s | ~7.4-8.9 (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.
→ All models on the NVIDIA GeForce RTX 5090
→ FragLlama: Next-fragment prediction for molecular design specs
→ Other hardware
FAQ
Can the NVIDIA GeForce RTX 5090 run FragLlama: Next-fragment prediction for molecular design?▶
Yes. At Q4 it generates ~333.7-397.7 tok/s and fits up to 183k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~7.4-8.9 tok/s.
How much context fits?▶
At Q4, up to about 183k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.