Cheapest box to run Llama-4-Maverick-17B-128E at Q6_K
Meta’s Llama-4-Maverick-17B-128E is 400B parameters, 17B 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.
Llama-4-Maverick-17B-128E at Q6_K needs 331GB.
Cheapest Config that fits: Mac Studio, M5 Ultra, 512GB. Fits, 173GB headroom. Est. 43.0 tok/s.
No price yet: Ships late October 2026. Apple has not published a price.
Collecting price history — 1 of 90 days. No timing verdict yet.
Offers — Mac Studio, M5 Ultra, 512GB, 1TB
Ships late October 2026. Apple has not published a price.
- Mac mini, M6, 16GB$1,099Won't fit−318GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−310GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−302GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−310GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−287GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−298GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−272GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−287GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−272GB headroom
- Won't fit−210GB headroom
- Won't fit−211GB headroom
- Won't fit−242GB headroom
- GeForce RTX 5090, 32GB$5,699.99+hostWon't fit−300GB headroom
- Won't fit−83GB headroom
- Fits173GB headroomest. 43.0 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −318GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −310GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −302GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −310GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −287GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −298GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −272GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −287GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −272GB | — |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Won't fit | −210GB | — |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Won't fit | −211GB | — |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Won't fit | −242GB | — |
| GeForce RTX 5090 | 32GB | $5,699.99+host | $178 | Won't fit | −300GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Won't fit | −83GB | — |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 173GB | 43.0 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 = 400B × 0.820 B/param (6.56 bits) + 0.197 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (17B active × 0.820 B/param) = 13.9 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 512GB runs
- DeepSeek-V3Q5_K_M, est. 22.9 tok/s
- Llama-4-Maverick-17B-128EQ8_0, est. 33.2 tok/s
- Qwen3-235B-A22BBF16, est. 13.6 tok/s
- Mistral-Large-2411BF16, est. 3.2 tok/s
- gpt-oss-120bBF16, est. 58.8 tok/s
- Llama-4-Scout-17B-16EBF16, est. 17.6 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 504GB of 512GB.