Cheapest box to run gpt-oss-20b at Q6_K
OpenAI’s gpt-oss-20b is 21B parameters, 3.6B 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.
gpt-oss-20b at Q6_K needs 19GB.
Cheapest Config that fits: Mac mini, M6, 32GB. Fits, 10GB headroom. Est. 28.8 tok/s.
Best offer today: $1,499 at Apple — in stock, checked 1 hour ago.
Price history: 47 days recorded.
Offers — Mac mini, M6, 32GB, 512GB
- Mac mini, M6, 16GB$1,099Won't fit−6GB headroom
- Mac mini, M6, 24GB$1,299Tight2GB headroomest. 28.8 tok/s
- Mac mini, M6, 32GB$1,499Fits10GB headroomest. 28.8 tok/s
- Mac mini, M5 Pro, 24GB$1,699Tight2GB headroomest. 52.0 tok/s
- Mac mini, M5 Pro, 48GB$2,499Fits25GB headroomest. 52.0 tok/s
- Mac Studio, M5 Max, 36GB$2,499Fits14GB headroomest. 77.9 tok/s
- Mac mini, M5 Pro, 64GB$2,899Fits40GB headroomest. 52.0 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostFits12GB headroomest. 304 tok/s
- Fits40GB headroomest. 92.5 tok/s
- Mac Studio, M5 Max, 48GB$3,099Fits25GB headroomest. 104 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits40GB headroomest. 104 tok/s
- Fits102GB headroomest. 46.2 tok/s
- Fits70GB headroomest. 139 tok/s
- Fits101GB headroomest. 92.5 tok/s
- Fits102GB headroomest. 46.2 tok/s
- Fits101GB headroomest. 104 tok/s
- Fits70GB headroomest. 203 tok/s
- Fits229GB headroomest. 139 tok/s
- Fits229GB headroomest. 203 tok/s
- Fits485GB headroomest. 203 tok/s
| Config | Memory | Price | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | Won't fit | −6GB | — |
| Mac mini, M6 | 24GB | $1,299 | Tight | 2GB | 28.8 tok/s |
| Mac mini, M6 | 32GB | $1,499 | Fits | 10GB | 28.8 tok/s |
| Mac mini, M5 Pro | 24GB | $1,699 | Tight | 2GB | 52.0 tok/s |
| Mac mini, M5 Pro | 48GB | $2,499 | Fits | 25GB | 52.0 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | Fits | 14GB | 77.9 tok/s |
| Mac mini, M5 Pro | 64GB | $2,899 | Fits | 40GB | 52.0 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | Fits | 12GB | 304 tok/s |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | Fits | 40GB | 92.5 tok/s |
| Mac Studio, M5 Max | 48GB | $3,099 | Fits | 25GB | 104 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | Fits | 40GB | 104 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | Fits | 102GB | 46.2 tok/s |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | Fits | 70GB | 139 tok/s |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | Fits | 101GB | 92.5 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | Fits | 102GB | 46.2 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | Fits | 101GB | 104 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | Fits | 70GB | 203 tok/s |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | Fits | 229GB | 139 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | Fits | 229GB | 203 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | Fits | 485GB | 203 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 = 21B × 0.820 B/param (6.56 bits) + 0.049 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (3.6B active × 0.820 B/param) = 3.0 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 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.