Cheapest box to run Qwen3-30B-A3B at Q8_0
Alibaba’s Qwen3-30B-A3B is 30.5B parameters, 3.3B of them read per token. 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.
Qwen3-30B-A3B at Q8_0 needs 34GB.
Cheapest Config that fits: Mac mini, M5 Pro, 48GB. Fits, 10GB headroom. Est. 43.8 tok/s.
Best offer today: $2,499 at Apple — in stock, checked 45 hours ago.
Collecting price history — 3 of 90 days. No timing verdict yet.
Offers — Mac mini, M5 Pro, 48GB, 512GB
- Applelist pricein stockchecked 45 hours ago$2,499
- Mac mini, M6, 16GB$1,099Won't fit−21GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−13GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−5GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−13GB headroom
- Mac mini, M5 Pro, 48GB$2,499Fits10GB headroomest. 43.8 tok/s
- Mac Studio, M5 Max, 36GB$2,499Won't fit−1GB headroom
- Mac mini, M5 Pro, 64GB$2,899Fits24GB headroomest. 43.8 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−4GB headroom
- Fits24GB headroomest. 77.9 tok/s
- Mac Studio, M5 Max, 48GB$3,099Fits10GB headroomest. 87.6 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits24GB headroomest. 87.6 tok/s
- Fits86GB headroomest. 38.9 tok/s
- Fits54GB headroomest. 117 tok/s
- Fits86GB headroomest. 77.9 tok/s
- Fits86GB headroomest. 38.9 tok/s
- Fits86GB headroomest. 87.6 tok/s
- Fits54GB headroomest. 171 tok/s
- Fits214GB headroomest. 117 tok/s
- Fits214GB headroomest. 171 tok/s
- Fits470GB headroomest. 171 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −21GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −13GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −5GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −13GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Fits | 10GB | 43.8 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −1GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Fits | 24GB | 43.8 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −4GB | — |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Fits | 24GB | 77.9 tok/s |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Fits | 10GB | 87.6 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Fits | 24GB | 87.6 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Fits | 86GB | 38.9 tok/s |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Fits | 54GB | 117 tok/s |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Fits | 86GB | 77.9 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 86GB | 38.9 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 86GB | 87.6 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 54GB | 171 tok/s |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Fits | 214GB | 117 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 214GB | 171 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 470GB | 171 tok/s |
GeForce RTX 5090 is a card, not a computer: every price shown for it is the cheapest listing we found plus $1,000 for the minimum host machine.
memory = 30.5B × 1.063 B/param (8.5 bits) + 0.098 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (3.3B active × 1.063 B/param) = 3.5 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
- Qwen3-30B-A3BQ8_0, est. 43.8 tok/s
- Gemma-3-27BQ8_0, est. 7.0 tok/s
- Devstral-Small-2507Q8_0, est. 7.8 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.