← CanIRunLocal · All hardware · All models
Yes. DeepSeek Coder V2 Lite 16B (MoE) fits in fast memory on M3 Max (40c GPU) 128GB at Q8_0 (18 GB needed, 96.0 GB usable), about 62 tok/s (estimate).
Hardware: M3 Max (40c GPU) 128GB · Model: DeepSeek Coder V2 Lite 16B (MoE)
M3 Max (40c GPU) 128GB has 128 GB of unified memory and 400 GB/s of memory bandwidth. The checker leaves macOS a share of unified memory, so Ollama can use 96.0 GB. DeepSeek Coder V2 Lite 16B (MoE) (15.7B, 2.4B active, DeepSeek) at Q8_0 is a 17 GB download and needs 18 GB with headroom for about 8k tokens.
| Best quant | Q8_0 |
|---|---|
| Download size | 17 GB |
| Memory needed | 18 GB |
| Usable fast memory | 96.0 GB |
| Fit | fits in fast memory |
| Estimated speed | ~62 tok/s (estimate) |
| Command | ollama run deepseek-coder-v2:16b-lite-instruct-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