Cheapest box to run gpt-oss-20b at BF16
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 BF16 needs 44GB.
Cheapest Config that fits: Mac mini, M5 Pro, 64GB. Fits, 15GB headroom. Est. 21.3 tok/s.
Best offer today: $2,899 at Apple — in stock, checked 10 hours ago.
Collecting price history — 2 of 90 days. No timing verdict yet.
Offers — Mac mini, M5 Pro, 64GB, 512GB
- Mac mini, M6, 16GB$1,099Won't fit−31GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−23GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−15GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−23GB headroom
- Mac mini, M5 Pro, 48GB$2,499Tight1GB headroomest. 21.3 tok/s
- Mac Studio, M5 Max, 36GB$2,499Won't fit−11GB headroom
- Mac mini, M5 Pro, 64GB$2,899Fits15GB headroomest. 21.3 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−13GB headroom
- Mac Studio, M5 Max, 48GB$3,099Tight1GB headroomest. 42.6 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits15GB headroomest. 42.6 tok/s
- Fits77GB headroomest. 19.0 tok/s
- Fits77GB headroomest. 19.0 tok/s
- Fits76GB headroomest. 42.6 tok/s
- Fits45GB headroomest. 83.3 tok/s
- Fits204GB headroomest. 83.3 tok/s
- Fits460GB headroomest. 83.3 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −31GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −23GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −15GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −23GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Tight | 1GB | 21.3 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −11GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Fits | 15GB | 21.3 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −13GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Tight | 1GB | 42.6 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Fits | 15GB | 42.6 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Fits | 77GB | 19.0 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 77GB | 19.0 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 76GB | 42.6 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 45GB | 83.3 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 204GB | 83.3 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 460GB | 83.3 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 × 2.000 B/param (16 bits) + 0.049 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (3.6B active × 2.000 B/param) = 7.2 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 64GB runs
- Llama-4-Scout-17B-16EQ3_K_M, est. 18.5 tok/s
- Llama-3.3-70BQ5_K_M, est. 4.0 tok/s
- Qwen3-32BQ8_0, est. 5.9 tok/s
- Gemma-3-27BBF16, est. 3.7 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 59GB of 64GB.