← CanIRunLocal · All hardware · All models
No. d3 27B (vLLM Semantic Router, decision) does not fit on M1 Max (24c GPU) 32GB: Q8_0 needs 32.3 GB and 24.0 GB is usable.
Hardware: M1 Max (24c GPU) 32GB · Model: d3 27B (vLLM Semantic Router, decision)
M1 Max (24c GPU) 32GB has 32 GB of unified memory and 400 GB/s of memory bandwidth. M1 Max (32c GPU) 32GB and M2 Max (30c GPU) 32GB and M2 Max (38c GPU) 32GB have the same memory and bandwidth, so the fit numbers match; this page is M1 Max (24c GPU) 32GB. The checker leaves macOS a share of unified memory, so Ollama can use 24.0 GB. d3 27B (vLLM Semantic Router, decision) (27B, vLLM Semantic Router) at Q8_0 is a 30 GB download and needs 32.3 GB with headroom for about 8k tokens.
| Best quant | Q8_0 |
|---|---|
| Download size | 30 GB |
| Memory needed | 32.3 GB |
| Usable fast memory | 24.0 GB |
| Fit | does not fit |
| Estimated speed | no estimate, it does not fit |
| Command | ollama run d3:27b-q8_0 |
Smallest hardware that runs it: RTX 3090 24GB at Q8_0 (partial offload (slower), 32.3 GB needed, 23.5 GB usable, about 4 tok/s (estimate)). Can a RTX 3090 24GB run d3 27B (vLLM Semantic Router, decision)?
A smaller vLLM Semantic Router model that fits this hardware: d3 9B (vLLM Semantic Router, decision) at Q8_0 (fits in fast memory, 11.4 GB). Can a M1 Max (24c GPU) 32GB run d3 9B (vLLM Semantic Router, decision)?
Open M1 Max (24c GPU) 32GB in the checkerEstimates, 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