← CanIRunLocal · All hardware · All models
Yes. Nimble 9B (Bespoke Labs, decision) fits in fast memory on RTX 5080 16GB at Q8_0 (10.6 GB needed, 15.5 GB usable), about 61 tok/s (estimate).
Hardware: RTX 5080 16GB · Model: Nimble 9B (Bespoke Labs, decision)
RTX 5080 16GB has 16 GB of VRAM and 960 GB/s of memory bandwidth. The checker holds back 0.5 GB of VRAM for the driver and assumes 16 GB of system RAM, so fast memory usable by Ollama is 15.5 GB. Partial offload may also use 60% of that system RAM. Nimble 9B (Bespoke Labs, decision) (9B, Bespoke Labs) at Q8_0 is a 9.5 GB download and needs 10.6 GB with headroom for about 8k tokens.
| Best quant | Q8_0 |
|---|---|
| Download size | 9.5 GB |
| Memory needed | 10.6 GB |
| Usable fast memory | 15.5 GB |
| Fit | fits in fast memory |
| Estimated speed | ~61 tok/s (estimate) |
| Command | ollama run nimble:9b-q8_0 |
Estimates, not benchmarks. Model sizes are ollama.com download sizes; memory needed adds headroom for ~8k tokens of context; speeds are rough figures from memory bandwidth. Data updated 2026-10-23.
A project by OnDevice Labs · llms.txt