← CanIRunLocal · All hardware · All models
Yes, with partial offload (slower). DeepSeek Coder V2 Lite 16B (MoE) on RTX 3060 Ti 8GB at Q4_K_M uses system RAM (11 GB needed, 7.5 GB usable), about 31 tok/s (estimate).
Hardware: RTX 3060 Ti 8GB · Model: DeepSeek Coder V2 Lite 16B (MoE)
RTX 3060 Ti 8GB has 8 GB of VRAM and 448 GB/s of memory bandwidth. RTX 3070 8GB has the same memory and bandwidth, so the fit numbers match; this page is RTX 3060 Ti 8GB. 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 7.5 GB. Partial offload may also use 60% of that system RAM. DeepSeek Coder V2 Lite 16B (MoE) (15.7B, 2.4B active, DeepSeek) at Q4_K_M is a 10 GB download and needs 11 GB with headroom for about 8k tokens.
| Best quant | Q4_K_M |
|---|---|
| Download size | 10 GB |
| Memory needed | 11 GB |
| Usable fast memory | 7.5 GB |
| Fit | partial offload (slower) |
| Estimated speed | ~31 tok/s (estimate) |
| Command | ollama run deepseek-coder-v2:16b-lite-instruct-q4_K_M |
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