Cheapest box to run MiniMax-M2 at Q3_K_M
MiniMax’s MiniMax-M2 is 229B parameters, 10B 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.
MiniMax-M2 at Q3_K_M needs 115GB.
Cheapest Config that fits: Mac Studio, M3 Ultra, 256GB, refurbished. Fits, 133GB headroom. Est. 83.8 tok/s.
Best offer today: $7,729 at Apple Certified Refurbished — in stock, checked 1 hour ago.
Collecting price history — 3 of 90 days. No timing verdict yet.
Offers — Mac Studio, M3 Ultra, 256GB, 1TB, refurbished
- Apple Certified RefurbishedRefurbished by Apple, with Apple's one-year limited warranty and 14-day returns. Apple says refurbished supply is usually very limited and often runs out, so this row can go dead faster than a new one.in stockchecked 1 hour ago$7,729
- Mac mini, M6, 16GB$1,099Won't fit−102GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−94GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−86GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−94GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−71GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−82GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−56GB headroom
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−84GB headroom
- Won't fit−56GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−71GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−56GB headroom
- Tight5GB headroomest. 27.9 tok/s
- Won't fit−27GB headroom
- Tight5GB headroomest. 55.9 tok/s
- Tight5GB headroomest. 27.9 tok/s
- Tight5GB headroomest. 62.8 tok/s
- Won't fit−27GB headroom
- Fits133GB headroomest. 83.8 tok/s
- Fits133GB headroomest. 123 tok/s
- Fits389GB headroomest. 123 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −102GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −94GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −86GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −94GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −71GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −82GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −56GB | — |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −84GB | — |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Won't fit | −56GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −71GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −56GB | — |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Tight | 5GB | 27.9 tok/s |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Won't fit | −27GB | — |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Tight | 5GB | 55.9 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Tight | 5GB | 27.9 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Tight | 5GB | 62.8 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Won't fit | −27GB | — |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Fits | 133GB | 83.8 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 133GB | 123 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 389GB | 123 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 = 229B × 0.489 B/param (3.91 bits) + 0.254 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (10B active × 0.489 B/param) = 4.9 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 M3 Ultra 256GB refurb runs
- Qwen3-Coder-480B-A35B-InstructQ3_K_M, est. 23.9 tok/s
- Llama-4-Maverick-17B-128EQ4_K_M, est. 39.9 tok/s
- GLM-4.6Q4_K_M, est. 21.2 tok/s
- Qwen3-235B-A22BQ6_K, est. 22.7 tok/s
- MiniMax-M2Q8_0, est. 38.5 tok/s
- Mistral-Large-2411Q8_0, est. 4.1 tok/s
Best quant that still fits at 8K context, largest model first. Full list on the Config page.
Bandwidth 819GB/s, usable memory 248GB of 256GB.