Cheapest box to run Gemma-4-31B at Q4_K_M
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 Q4_K_M needs 21GB.
Cheapest Config that fits: Mac mini, M6, 32GB. Fits, 8GB headroom. Est. 6.0 tok/s.
Best offer today: $1,499 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, M6, 32GB, 512GB
- Mac mini, M6, 16GB$1,099Won't fit−8GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−0GB headroom
- Mac mini, M6, 32GB$1,499Fits8GB headroomest. 6.0 tok/s
- Mac mini, M5 Pro, 24GB$1,699Won't fit−0GB headroom
- Mac mini, M5 Pro, 48GB$2,499Fits23GB headroomest. 10.8 tok/s
- Mac Studio, M5 Max, 36GB$2,499Fits12GB headroomest. 16.1 tok/s
- Mac mini, M5 Pro, 64GB$2,899Fits38GB headroomest. 10.8 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostFits10GB headroomest. 62.8 tok/s
- Fits38GB headroomest. 19.1 tok/s
- Mac Studio, M5 Max, 48GB$3,099Fits23GB headroomest. 21.5 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits38GB headroomest. 21.5 tok/s
- Fits99GB headroomest. 9.6 tok/s
- Fits67GB headroomest. 28.7 tok/s
- Fits99GB headroomest. 19.1 tok/s
- Fits99GB headroomest. 9.6 tok/s
- Fits99GB headroomest. 21.5 tok/s
- Fits67GB headroomest. 42.1 tok/s
- Fits227GB headroomest. 28.7 tok/s
- Fits227GB headroomest. 42.1 tok/s
- Fits483GB headroomest. 42.1 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −8GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −0GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Fits | 8GB | 6.0 tok/s |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −0GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Fits | 23GB | 10.8 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Fits | 12GB | 16.1 tok/s |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Fits | 38GB | 10.8 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Fits | 10GB | 62.8 tok/s |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Fits | 38GB | 19.1 tok/s |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Fits | 23GB | 21.5 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Fits | 38GB | 21.5 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Fits | 99GB | 9.6 tok/s |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Fits | 67GB | 28.7 tok/s |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Fits | 99GB | 19.1 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 99GB | 9.6 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 99GB | 21.5 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 67GB | 42.1 tok/s |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Fits | 227GB | 28.7 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 227GB | 42.1 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 483GB | 42.1 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 × 0.604 B/param (4.83 bits) + 0.168 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.65 ÷ (30.7B active × 0.604 B/param) = 18.5 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 M6 32GB runs
- Qwen3.5-35B-A3BQ5_K_M, est. 40.0 tok/s
- Qwen3-32BQ5_K_M, est. 4.9 tok/s
- Gemma-4-31BQ6_K, est. 4.4 tok/s
- Qwen3-30B-A3BQ6_K, est. 31.4 tok/s
- Gemma-3-27BQ6_K, est. 5.0 tok/s
- Qwen3.6-27BQ6_K, est. 5.0 tok/s
Best quant that still fits at 8K context, largest model first. Full list on the Config page.
Bandwidth 170GB/s, usable memory 29GB of 32GB.