Mac mini, M6, 32GB vs GeForce RTX 5090, 32GB
Both hold Gemma-3-27B at Q6_K. GeForce RTX 5090, 32GB decodes an est. 52.6 tok/s against 5.0, and costs $1,500.99 more — $31.52.03628948117239 per extra token per second. Worth it if you read the output as it arrives; not if you queue jobs.
- Fit
- Tight
- Fits
- Headroom
- 4GB
- 5GB
- Est. speed
- 5.0 tok/s
- 52.6 tok/s
- Cheapest today
- $1,499 at Apple
- $2,999.99 at Best Buy
- Per GB of memory
- $47
- $94
- Memory
- 32GB unified memory
- 32GB GDDR7 memory
- Usable for a model
- 29GB
- 31GB
- Memory bandwidth
- 170GB/s
- 1.8TB/s
- Storage
- 512GB
- —
- Lowest recorded
- $1,499 in 2d
- $2,999 in 1d
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 fitWon't fit
- gpt-oss-120bWon't fitWon't fit
- Mistral-Large-2411Won't fitWon't fit
- Llama-4-Scout-17B-16EWon't fitWon't fit
- Qwen3-32BWon't fitTight
- Gemma-3-27BTightFits
- gpt-oss-20bFitsFits
- Llama-4-Maverick-17B-128EWon't fitWon't fit
- Llama-4-BehemothWon't fitWon't fit
Every model we track at Q6_K. The rows where the two boxes agree are dimmed, because they are not what you are choosing between.