Can Mac mini M2 Pro 32 GB run Qwen3.8 27B?
32 GB unified · 200 GB/s memory bandwidth · Qwen3.8 27B has 27.8B parameters.
Short answer
Yes, acceptable for chat; not recommended for agentic coding.
Best quantization for chat: 4-bit (15 GB of weights) on LM Studio. Generation: 8–11 tok/s, leaving about 8.0 GB free for your other apps.
How it rates for each workload
Same model and machine, different jobs. Agents and long documents send far bigger prompts than chat, so they need much more speed and memory.
| Workload | Verdict | Quant | Generation | First reply | Context | |
|---|---|---|---|---|---|---|
| Chatwith Open WebUI | Acceptable | 4-bit | 8–11 tok/s | ≈9 s | 8K | Details |
| Coding questionswith Continue | Acceptable | 4-bit | 8–11 tok/s | ≈55 s | 16K | Details |
| Coding in a repositorywith Aider | Borderline | 4-bit | 7–10 tok/s | ≈3 min | 32K | Details |
| Agentic codingwith OpenCode | Technically Runs | 5-bit | 6–8 tok/s | ≈4 min | 32K | Details |
| Long documents (64K)with Open WebUI | Borderline | 4-bit | 7–9 tok/s | ≈12 min | 64K | Details |
Speeds are estimates calibrated against measured benchmarks on similar hardware. “First reply” is the time to read the workload’s opening prompt. How we calculate
Get it running
Copy-paste commands for the 4-bit build with a 16K context. Ollama and llama.cpp download the file straight from Hugging Face.
ollama run hf.co/unsloth/Qwen3.8-27B-GGUF:UD-Q4_K_XL
Downloads 17.56 GB (UD-Q4_K_XL). Ollama's default context is small: set OLLAMA_CONTEXT_LENGTH=16384 before `ollama serve` for long prompts.
Other models for Mac mini M2 Pro 32 GB
The best-rated models for chat on this machine.
- Qwen3.6 35B-A3B55–75 tok/sExcellent
- Gemma 4 26B-A4B45–60 tok/sExcellent
- LFM2.5 24B-A2B75–100 tok/sExcellent
- gpt-oss-20b45–60 tok/sExcellent
- Qwen3.5 35B-A3B55–75 tok/sExcellent
- GLM-4.7-Flash (30B-A3B)50–70 tok/sExcellent
Cheapest machines for Qwen3.8 27B
Hardware rated comfortable or better for chat, lowest price first.
- Desktop PC · Intel Arc Pro B60 24GB · 32 GB RAM$1,500 · 11–18 tok/sComfortable
- Desktop PC · Radeon RX 7900 XT · 32 GB RAM$1,550 · 25–40 tok/sExcellent
- Ryzen AI Max+ 395 mini PC / Framework Desktop · 64 GB$1,600 · 12–16 tok/sComfortable
- Desktop PC · Intel Arc Pro B60 24GB · 64 GB RAM$1,628 · 11–18 tok/sComfortable
- Desktop PC · GeForce RTX 3090 · 32 GB RAM$1,700 · 30–40 tok/sExcellent
- Mac mini M4 Pro 48 GB$1,799 · 11–14 tok/sComfortable
Same chip, other memory: 16 GB
Different tool, context size or workload?
These pages use typical settings. The full check models your exact setup.