Cheapest box to run Qwen3-Coder-480B-A35B-Instruct at Q6_K
Alibaba’s Qwen3-Coder-480B-A35B-Instruct is 480B parameters, 35B 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.
Qwen3-Coder-480B-A35B-Instruct at Q6_K needs 397GB.
Cheapest Config that fits: Mac Studio, M5 Ultra, 512GB. Fits, 107GB headroom. Est. 20.9 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−384GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−376GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−368GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−376GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−353GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−364GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−338GB headroom
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−366GB headroom
- Won't fit−338GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−353GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−338GB headroom
- Won't fit−277GB headroom
- Won't fit−309GB headroom
- Won't fit−277GB headroom
- Won't fit−277GB headroom
- Won't fit−277GB headroom
- Won't fit−309GB headroom
- Won't fit−149GB headroom
- Won't fit−149GB headroom
- Fits107GB headroomest. 20.9 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −384GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −376GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −368GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −376GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −353GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −364GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −338GB | — |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −366GB | — |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Won't fit | −338GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −353GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −338GB | — |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Won't fit | −277GB | — |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Won't fit | −309GB | — |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Won't fit | −277GB | — |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Won't fit | −277GB | — |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Won't fit | −277GB | — |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Won't fit | −309GB | — |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Won't fit | −149GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Won't fit | −149GB | — |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 107GB | 20.9 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 = 480B × 0.820 B/param (6.56 bits) + 0.254 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (35B active × 0.820 B/param) = 28.7 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.