Cheapest box to run GLM-4.6 at Q5_K_M
Z.ai’s GLM-4.6 is 357B parameters, 32B 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.
GLM-4.6 at Q5_K_M needs 257GB.
Cheapest Config that fits: Mac Studio, M5 Ultra, 512GB. Fits, 247GB headroom. Est. 26.5 tok/s.
No price yet: Ships late October 2026. Apple has not published a price.
Collecting price history — 3 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−244GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−236GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−228GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−236GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−213GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−224GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−198GB headroom
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−227GB headroom
- Won't fit−198GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−213GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−198GB headroom
- Won't fit−137GB headroom
- Won't fit−169GB headroom
- Won't fit−137GB headroom
- Won't fit−137GB headroom
- Won't fit−137GB headroom
- Won't fit−169GB headroom
- Won't fit−9GB headroom
- Won't fit−9GB headroom
- Fits247GB headroomest. 26.5 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −244GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −236GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −228GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −236GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −213GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −224GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −198GB | — |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −227GB | — |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Won't fit | −198GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −213GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −198GB | — |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Won't fit | −137GB | — |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Won't fit | −169GB | — |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Won't fit | −137GB | — |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Won't fit | −137GB | — |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Won't fit | −137GB | — |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Won't fit | −169GB | — |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Won't fit | −9GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Won't fit | −9GB | — |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 247GB | 26.5 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 = 357B × 0.709 B/param (5.67 bits) + 0.377 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (32B active × 0.709 B/param) = 22.7 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
- Kimi-K2-InstructQ3_K_M, est. 38.4 tok/s
- DeepSeek-V3Q5_K_M, est. 22.9 tok/s
- DeepSeek-R1Q5_K_M, est. 22.9 tok/s
- Qwen3-Coder-480B-A35B-InstructQ6_K, est. 20.9 tok/s
- Llama-4-Maverick-17B-128EQ8_0, est. 33.2 tok/s
- GLM-4.6Q8_0, 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.