Cheapest box to run gpt-oss-120b at Q8_0
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 Q8_0 needs 126GB.
Cheapest Config that fits: Mac Studio, M5 Ultra, 256GB. Fits, 122GB headroom. Est. 111 tok/s.
Best offer today: $9,499 at Apple — in stock, checked 9 hours ago.
Collecting price history — 1 of 90 days. No timing verdict yet.
Offers — Mac Studio, M5 Ultra, 256GB, 1TB
- Applelist pricein stockchecked 9 hours ago$9,499
- Mac mini, M6, 16GB$1,099Won't fit−113GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−105GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−97GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−105GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−82GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−93GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−67GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−82GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−67GB headroom
- Won't fit−6GB headroom
- Won't fit−6GB headroom
- Won't fit−38GB headroom
- GeForce RTX 5090, 32GB$5,699.99+hostWon't fit−95GB headroom
- Fits122GB headroomest. 111 tok/s
- Fits378GB headroomest. 111 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −113GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −105GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −97GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −105GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −82GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −93GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −67GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −82GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −67GB | — |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Won't fit | −6GB | — |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Won't fit | −6GB | — |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Won't fit | −38GB | — |
| GeForce RTX 5090 | 32GB | $5,699.99+host | $178 | Won't fit | −95GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 122GB | 111 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 378GB | 111 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 × 1.063 B/param (8.5 bits) + 0.074 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (5.1B active × 1.063 B/param) = 5.4 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 Studio M5 Ultra 256GB runs
- Llama-4-Maverick-17B-128EQ4_K_M, est. 58.5 tok/s
- Qwen3-235B-A22BQ6_K, est. 33.3 tok/s
- Mistral-Large-2411Q8_0, est. 6.0 tok/s
- gpt-oss-120bBF16, est. 58.8 tok/s
- Llama-4-Scout-17B-16EBF16, est. 17.6 tok/s
- Llama-3.3-70BBF16, est. 5.6 tok/s
Best quant that still fits at 8K context, largest model first. Full list on the Config page.
Bandwidth 1.2TB/s, usable memory 248GB of 256GB.