Cheapest box to run Gemma-4-12B at Q6_K
Google’s Gemma-4-12B is 11.95B 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-4-12B at Q6_K needs 12GB.
Cheapest Config that fits: Mac mini, M6, 24GB. Fits, 9GB headroom. Est. 11.3 tok/s.
Best offer today: $1,299 at Best Buy — in stock, checked 33 days ago.
Price history: 34 days recorded.
Dense. 8 of 48 layers are global attention; the sliding-window layers stop growing past 1024 tokens.
Offers — Mac mini, M6, 24GB, 512GB
- Mac mini, M6, 16GB$1,099Tight1GB headroomest. 11.3 tok/s
- Mac mini, M6, 24GB$1,299Fits9GB headroomest. 11.3 tok/s
- Mac mini, M6, 32GB$1,499Fits17GB headroomest. 11.3 tok/s
- Mac mini, M5 Pro, 24GB$1,699Fits9GB headroomest. 20.4 tok/s
- Mac mini, M5 Pro, 48GB$2,499Fits33GB headroomest. 20.4 tok/s
- Mac Studio, M5 Max, 36GB$2,499Fits21GB headroomest. 30.5 tok/s
- Mac mini, M5 Pro, 64GB$2,899Fits47GB headroomest. 20.4 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostFits19GB headroomest. 119 tok/s
- Fits47GB headroomest. 36.2 tok/s
- Mac Studio, M5 Max, 48GB$3,099Fits33GB headroomest. 40.7 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits47GB headroomest. 40.7 tok/s
- Fits109GB headroomest. 18.1 tok/s
- Fits77GB headroomest. 54.3 tok/s
- Fits108GB headroomest. 36.2 tok/s
- Fits109GB headroomest. 18.1 tok/s
- Fits108GB headroomest. 40.7 tok/s
- Fits77GB headroomest. 79.6 tok/s
- Fits236GB headroomest. 54.3 tok/s
- Fits236GB headroomest. 79.6 tok/s
- Fits492GB headroomest. 79.6 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Tight | 1GB | 11.3 tok/s |
| Mac mini, M6 | 24GB | $1,299 | $54 | Fits | 9GB | 11.3 tok/s |
| Mac mini, M6 | 32GB | $1,499 | $47 | Fits | 17GB | 11.3 tok/s |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Fits | 9GB | 20.4 tok/s |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Fits | 33GB | 20.4 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Fits | 21GB | 30.5 tok/s |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Fits | 47GB | 20.4 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Fits | 19GB | 119 tok/s |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Fits | 47GB | 36.2 tok/s |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Fits | 33GB | 40.7 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Fits | 47GB | 40.7 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Fits | 109GB | 18.1 tok/s |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Fits | 77GB | 54.3 tok/s |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Fits | 108GB | 36.2 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 109GB | 18.1 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 108GB | 40.7 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 77GB | 79.6 tok/s |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Fits | 236GB | 54.3 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 236GB | 79.6 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 492GB | 79.6 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 = 11.95B × 0.820 B/param (6.56 bits) + 0.067 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.65 ÷ (11.95B active × 0.820 B/param) = 9.8 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 24GB runs
- Qwen3.5-35B-A3BQ3_K_M, est. 58.0 tok/s
- Qwen3-32BQ3_K_M, est. 7.1 tok/s
- Gemma-4-31BQ3_K_M, est. 7.4 tok/s
- Qwen3-30B-A3BQ4_K_M, est. 42.7 tok/s
- Gemma-3-27BQ4_K_M, est. 6.8 tok/s
- Qwen3.6-27BQ5_K_M, est. 5.8 tok/s
Best quant that still fits at 8K context, largest model first. Full list on the Config page.
Bandwidth 170GB/s, usable memory 21GB of 24GB.