Cheapest box to run gpt-oss-120b at Q3_K_M
OpenAI’s gpt-oss-120b is 117B parameters, 5.1B 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-120b at Q3_K_M needs 59GB.
Cheapest Config that fits: DGX Spark, GB10 Grace Blackwell, 128GB. Fits, 61GB headroom. Est. 54.8 tok/s.
Best offer today: $4,699 at NVIDIA — in stock, checked 9 hours ago.
Collecting price history — 1 of 90 days. No timing verdict yet.
Offers — DGX Spark, GB10 Grace Blackwell, 128GB, 4TB
- NVIDIAlist pricein stockchecked 9 hours ago$4,699
- Mac mini, M6, 16GB$1,099Won't fit−46GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−38GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−30GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−38GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−15GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−26GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−0GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−15GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−0GB headroom
- Fits61GB headroomest. 54.8 tok/s
- Fits61GB headroomest. 123 tok/s
- Fits29GB headroomest. 241 tok/s
- GeForce RTX 5090, 32GB$5,699.99+hostWon't fit−28GB headroom
- Fits189GB headroomest. 241 tok/s
- Fits445GB headroomest. 241 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −46GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −38GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −30GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −38GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −15GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −26GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −0GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −15GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −0GB | — |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 61GB | 54.8 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 61GB | 123 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 29GB | 241 tok/s |
| GeForce RTX 5090 | 32GB | $5,699.99+host | $178 | Won't fit | −28GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 189GB | 241 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 445GB | 241 tok/s |
GeForce RTX 5090 is a card, not a computer: its price is the US street median plus $1,000 for the minimum host machine. NVIDIA's $1,999 MSRP is not a price anyone pays.
memory = 117B × 0.489 B/param (3.91 bits) + 0.074 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (5.1B active × 0.489 B/param) = 2.5 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 DGX Spark GB10 Grace Blackwell 128GB runs
- Qwen3-235B-A22BQ3_K_M, est. 12.7 tok/s
- Mistral-Large-2411Q6_K, est. 1.8 tok/s
- gpt-oss-120bQ6_K, est. 32.6 tok/s
- Llama-4-Scout-17B-16EQ8_0, est. 7.6 tok/s
- Llama-3.3-70BQ8_0, est. 2.4 tok/s
- Qwen3-32BBF16, est. 2.8 tok/s
Best quant that still fits at 8K context, largest model first. Full list on the Config page.
Bandwidth 273GB/s, usable memory 120GB of 128GB.