Cheapest box to run Llama-4-Maverick-17B-128E at Q3_K_M
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 Q3_K_M needs 198GB.
Cheapest Config that fits: Mac Studio, M5 Ultra, 256GB. Fits, 50GB headroom. Est. 72.2 tok/s.
Best offer today: $9,499 at Apple — in stock, checked 10 hours ago.
Collecting price history — 2 of 90 days. No timing verdict yet.
Offers — Mac Studio, M5 Ultra, 256GB, 1TB
- Mac mini, M6, 16GB$1,099Won't fit−185GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−177GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−169GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−177GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−154GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−165GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−139GB headroom
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−168GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−154GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−139GB headroom
- Won't fit−78GB headroom
- Won't fit−78GB headroom
- Won't fit−78GB headroom
- Won't fit−110GB headroom
- Fits50GB headroomest. 72.2 tok/s
- Fits306GB headroomest. 72.2 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −185GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −177GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −169GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −177GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −154GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −165GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −139GB | — |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −168GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −154GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −139GB | — |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Won't fit | −78GB | — |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Won't fit | −78GB | — |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Won't fit | −78GB | — |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Won't fit | −110GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 50GB | 72.2 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 306GB | 72.2 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 = 400B × 0.489 B/param (3.91 bits) + 0.197 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (17B active × 0.489 B/param) = 8.3 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.