Cheapest box to run Gemma-4-31B at Q6_K
Google’s Gemma-4-31B is 30.7B 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.
Gemma-4-31B at Q6_K needs 28GB.
Cheapest Config that fits: Mac mini, M5 Pro, 48GB. Fits, 16GB headroom. Est. 7.9 tok/s.
Best offer today: $2,499 at Apple — in stock, checked 32 minutes ago.
Price history: 34 days recorded.
Dense. Only 10 of 60 layers are global attention; the sliding-window layers stop growing past 1024 tokens, so the K/V figure counts the global layers alone.
Offers — Mac mini, M5 Pro, 48GB, 512GB
- Applelist pricein stockchecked 32 minutes ago$2,499
- Mac mini, M6, 16GB$1,099Won't fit−15GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−7GB headroom
- Mac mini, M6, 32GB$1,499Tight1GB headroomest. 4.4 tok/s
- Mac mini, M5 Pro, 24GB$1,699Won't fit−7GB headroom
- Mac mini, M5 Pro, 48GB$2,499Fits16GB headroomest. 7.9 tok/s
- Mac Studio, M5 Max, 36GB$2,499Fits5GB headroomest. 11.9 tok/s
- Mac mini, M5 Pro, 64GB$2,899Fits31GB headroomest. 7.9 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostTight3GB headroomest. 46.3 tok/s
- Fits31GB headroomest. 14.1 tok/s
- Mac Studio, M5 Max, 48GB$3,099Fits16GB headroomest. 15.9 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits31GB headroomest. 15.9 tok/s
- Fits93GB headroomest. 7.0 tok/s
- Fits61GB headroomest. 21.1 tok/s
- Fits92GB headroomest. 14.1 tok/s
- Fits93GB headroomest. 7.0 tok/s
- Fits92GB headroomest. 15.9 tok/s
- Fits61GB headroomest. 31.0 tok/s
- Fits220GB headroomest. 21.1 tok/s
- Fits220GB headroomest. 31.0 tok/s
- Fits476GB headroomest. 31.0 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −15GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −7GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Tight | 1GB | 4.4 tok/s |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −7GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Fits | 16GB | 7.9 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Fits | 5GB | 11.9 tok/s |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Fits | 31GB | 7.9 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Tight | 3GB | 46.3 tok/s |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Fits | 31GB | 14.1 tok/s |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Fits | 16GB | 15.9 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Fits | 31GB | 15.9 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Fits | 93GB | 7.0 tok/s |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Fits | 61GB | 21.1 tok/s |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Fits | 92GB | 14.1 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 93GB | 7.0 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 92GB | 15.9 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 61GB | 31.0 tok/s |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Fits | 220GB | 21.1 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 220GB | 31.0 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 476GB | 31.0 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 = 30.7B × 0.820 B/param (6.56 bits) + 0.168 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.65 ÷ (30.7B active × 0.820 B/param) = 25.2 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
- Qwen3-Coder-NextQ3_K_M, est. 105 tok/s
- Llama-3.3-70BQ3_K_M, est. 5.8 tok/s
- Qwen3.5-35B-A3BQ8_0, est. 48.2 tok/s
- Qwen3-32BQ8_0, est. 5.9 tok/s
- Gemma-4-31BQ8_0, est. 6.1 tok/s
- Qwen3-30B-A3BQ8_0, est. 43.8 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.