Can I run it?
Can I run Swift Qwen3.8 27B on Apple M5?
Heavily degraded
Heavily degraded fully in unified memory on your Apple M5 at IQ2_S, 8K context, about 8.1–11 tok/s. Weights 9.22 GB plus 0.57 GB of KV cache — 10.2 GB of 11.0 GB usable. Up to 16K context still fits.
Best quantIQ2_S
Working context8K
Max that fits16K
Speed8.1–11 tok/sEstimate from memory bandwidth, not a benchmark. Method.
Weights9.22 GB
KV cache0.57 GB
Total10.2 GB
Usable11.0 GB
Swift Qwen3.8 27B details · Apple M5 page · How this estimate is made
Run commands
llama.cpp
llama-server -hf ukisai/Swift-Qwen3.8-27B-GGUF:IQ2_S -c 8192 -ngl 64
No matching name in the Ollama library snapshot.
LM Studio
lms get ukisai/Swift-Qwen3.8-27B-GGUF # Then load quant IQ2_S
Assumptions
- Estimate from memory bandwidth, not a benchmark. Decode speed uses 55–75% of peak bandwidth.
- KV cache is FP16. Engine overhead assumes llama.cpp (0.3 GB + 1% of the weights). Ollama is closer to 0.9 GB + 3%, LM Studio closer to 1 GB + 3%.
- IQ2_S size is the real GGUF file size from Hugging Face.
- Unified memory uses 75% for the GPU by default, and always leaves about 5 GB for the OS.
- Speed is quoted at 8K context, not at the maximum context that still fits.
- This quant is IQ1 or IQ2. It can load, but quality is heavily degraded, so it is left off best-fit lists.
Larger models that also fit
- Can I run Qwen3.8 27B on Apple M5? — Tight fit, UD-Q2_K_XL, 8.1–11 tok/s (estimate)
- Can I run Qwen3.6 35B A3B on Apple M5? — Heavily degraded, UD-IQ2_XXS, 19–25 tok/s (estimate)
- Can I run Qwen3.5 35B A3B on Apple M5? — Heavily degraded, UD-IQ2_XXS, 19–25 tok/s (estimate)
Updated 2026-10-24. Profile on this page is the device default.