🔌
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 SeqVec?
Yes — here's how fast and up to what context, with real measured numbers.
SeqVec on the NVIDIA GeForce RTX 2050
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 73k | 439.6-523.9 t/s | ~39.2-46.8 (slow) |
| Q8 | yes | 75k | 879.1-1047.7 t/s | ~78.5-93.5 (slow) |
| Q4 | yes | 75k | 1569.9-1871 t/s | ~140.2-167.1 (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 SeqVec?▶
Yes. At Q4 it generates ~1569.9-1871 tok/s and fits up to 75k context; beyond that it spills to SSD swap and slows to ~140.2-167.1 tok/s.
How much context fits?▶
At Q4, up to about 75k tokens stay in fast memory; longer context spills to SSD swap and throughput collapses.