Mac mini, M6, 24GB vs Mac mini, M5 Pro, 48GB
Both hold Qwen3-32B at Q3_K_M. Mac mini, M5 Pro, 48GB decodes an est. 12.8 tok/s against 7.1, and costs $1,200 more — $210.75.8001122964597 per extra token per second. Worth it if you read the output as it arrives; not if you queue jobs.
- Fit
- Tight
- Fits
- Headroom
- 2GB
- 25GB
- Est. speed
- 7.1 tok/s
- 12.8 tok/s
- Cheapest today
- $1,299 at Apple
- $2,499 at Apple
- Per GB of memory
- $54
- $52
- Memory
- 24GB unified memory
- 48GB unified memory
- Usable for a model
- 21GB
- 44GB
- Memory bandwidth
- 170GB/s
- 307GB/s
- Storage
- 512GB
- 512GB
- Lowest recorded
- $1,299 in 2d
- $2,499 in 2d
Fit, headroom and speed are computed from each vendor’s published bandwidth and the model’s own size at 8K context. Prices are the cheapest listing we could read on each seller’s page. The arithmetic is here.
Where they actually differ
- Qwen3-235B-A22BWon't fitWon't fit
- DeepSeek-V3Won't fitWon't fit
- Llama-3.3-70BWon't fitFits
- gpt-oss-120bWon't fitWon't fit
- Mistral-Large-2411Won't fitWon't fit
- Llama-4-Scout-17B-16EWon't fitWon't fit
- Qwen3-32BTightFits
- Gemma-3-27BFitsFits
- gpt-oss-20bFitsFits
- Llama-4-Maverick-17B-128EWon't fitWon't fit
- Llama-4-BehemothWon't fitWon't fit
Every model we track at Q3_K_M. The rows where the two boxes agree are dimmed, because they are not what you are choosing between.