🔌
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 2050 run OPT-1.3B?
Yes — here's how fast and up to what context, with real measured numbers.
OPT-1.3B on the NVIDIA GeForce RTX 2050
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 37k | 31.4-37.5 t/s | ~2.8-3.3 (slow) |
| Q8 | yes | 56k | 62.9-75 t/s | ~5.6-6.7 (slow) |
| Q4 | yes | 65k | 112.3-133.8 t/s | ~10-12 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to SSD swap and speed drops to the "beyond" figure.
FAQ
Can the NVIDIA GeForce RTX 2050 run OPT-1.3B?▶
Yes. At Q4 it generates ~112.3-133.8 tok/s and fits up to 65k context; beyond that it spills to SSD swap and slows to ~10-12 tok/s.
How much context fits?▶
At Q4, up to about 65k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.