🔌
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 3080 10GB run ALiBi (L=3072, Lvalid = 3072)?
Yes — here's how fast and up to what context, with real measured numbers.
ALiBi (L=3072, Lvalid = 3072) on the NVIDIA GeForce RTX 3080 10GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 78k | 213.4-254.3 t/s | ~11.2-13.4 (slow) |
| Q8 | yes | 98k | 426.8-508.6 t/s | ~22.5-26.8 (slow) |
| Q4 | yes | 106k | 762.1-908.2 t/s | ~40.1-47.8 (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 3080 10GB
→ ALiBi (L=3072, Lvalid = 3072) specs
→ Other hardware
FAQ
Can the NVIDIA GeForce RTX 3080 10GB run ALiBi (L=3072, Lvalid = 3072)?▶
Yes. At Q4 it generates ~762.1-908.2 tok/s and fits up to 106k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~40.1-47.8 tok/s.
How much context fits?▶
At Q4, up to about 106k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.