Cheapest box to run Qwen3-32B at Q8_0
Alibaba’s Qwen3-32B is 32B parameters, dense. The number below is computed from that and each vendor’s published specs, not asserted. Switch the quant or the context and it recomputes in place.
Qwen3-32B at Q8_0 needs 37GB.
Cheapest Config that fits: Mac mini, M5 Pro, 48GB. Fits, 7GB headroom. Est. 5.9 tok/s.
Best offer today: $2,499 at Apple — in stock, checked 9 hours ago.
Collecting price history — 1 of 90 days. No timing verdict yet.
Offers — Mac mini, M5 Pro, 48GB, 512GB
- Applelist pricein stockchecked 9 hours ago$2,499
- Mac mini, M6, 16GB$1,099Won't fit−24GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−16GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−8GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−16GB headroom
- Mac mini, M5 Pro, 48GB$2,499Fits7GB headroomest. 5.9 tok/s
- Mac Studio, M5 Max, 36GB$2,499Won't fit−4GB headroom
- Mac mini, M5 Pro, 64GB$2,899Fits22GB headroomest. 5.9 tok/s
- Mac Studio, M5 Max, 48GB$3,099Fits7GB headroomest. 11.7 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits22GB headroomest. 11.7 tok/s
- Fits83GB headroomest. 5.2 tok/s
- Fits83GB headroomest. 11.7 tok/s
- Fits51GB headroomest. 22.9 tok/s
- GeForce RTX 5090, 32GB$5,699.99+hostWon't fit−7GB headroom
- Fits211GB headroomest. 22.9 tok/s
- Fits467GB headroomest. 22.9 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −24GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −16GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −8GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −16GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Fits | 7GB | 5.9 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −4GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Fits | 22GB | 5.9 tok/s |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Fits | 7GB | 11.7 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Fits | 22GB | 11.7 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 83GB | 5.2 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 83GB | 11.7 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 51GB | 22.9 tok/s |
| GeForce RTX 5090 | 32GB | $5,699.99+host | $178 | Won't fit | −7GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 211GB | 22.9 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 467GB | 22.9 tok/s |
GeForce RTX 5090 is a card, not a computer: its price is the US street median plus $1,000 for the minimum host machine. NVIDIA's $1,999 MSRP is not a price anyone pays.
memory = 32B × 1.063 B/param (8.5 bits) + 0.262 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.65 ÷ (32B active × 1.063 B/param) = 34.0 GB read per token
usable = memory − operating system reserve (8% on a Mac, 6% on a DGX Spark, 4% on a card)
Bandwidth is the vendor’s published figure. Every speed is an estimate and says so. Methodology.
What else Mac mini M5 Pro 48GB runs
- Llama-3.3-70BQ3_K_M, est. 5.8 tok/s
- Qwen3-32BQ8_0, est. 5.9 tok/s
- Gemma-3-27BQ8_0, est. 7.0 tok/s
- gpt-oss-20bBF16, est. 21.3 tok/s
Best quant that still fits at 8K context, largest model first. Full list on the Config page.
Bandwidth 307GB/s, usable memory 44GB of 48GB.