Cheapest box to run Llama-4-Scout-17B-16E at Q3_K_M
Meta’s Llama-4-Scout-17B-16E is 109B 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-Scout-17B-16E at Q3_K_M needs 56GB.
Cheapest Config that fits: Ascent GX10, GB10 Grace Blackwell, 128GB. Fits, 64GB headroom. Est. 16.4 tok/s.
Best offer today: $3,999.99 at Best Buy — in stock, checked 10 hours ago.
Collecting price history — 2 of 90 days. No timing verdict yet.
Offers — Ascent GX10, GB10 Grace Blackwell, 128GB, 1TB
- Mac mini, M6, 16GB$1,099Won't fit−43GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−35GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−27GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−35GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−12GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−23GB headroom
- Mac mini, M5 Pro, 64GB$2,899Tight3GB headroomest. 18.5 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−25GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−12GB headroom
- Mac Studio, M5 Max, 64GB$3,499Tight3GB headroomest. 36.9 tok/s
- Fits64GB headroomest. 16.4 tok/s
- Fits64GB headroomest. 16.4 tok/s
- Fits64GB headroomest. 36.9 tok/s
- Fits32GB headroomest. 72.2 tok/s
- Fits192GB headroomest. 72.2 tok/s
- Fits448GB headroomest. 72.2 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −43GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −35GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −27GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −35GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −12GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −23GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Tight | 3GB | 18.5 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −25GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −12GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Tight | 3GB | 36.9 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Fits | 64GB | 16.4 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 64GB | 16.4 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 64GB | 36.9 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 32GB | 72.2 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 192GB | 72.2 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 448GB | 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 = 109B × 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 Ascent GX10 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.