Cheapest box to run Qwen3-Coder-480B-A35B-Instruct at BF16
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 BF16 needs 963GB.
Nothing on the Grid runs Qwen3-Coder-480B-A35B-Instruct at BF16 today. The largest Config, Mac Studio, M5 Ultra, 512GB, is 459GB short. At it fits with 266GB headroom.
- Mac mini, M6, 16GB$1,099Won't fit−950GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−942GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−934GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−942GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−919GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−930GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−904GB headroom
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−933GB headroom
- Won't fit−904GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−919GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−904GB headroom
- Won't fit−843GB headroom
- Won't fit−875GB headroom
- Won't fit−843GB headroom
- Won't fit−843GB headroom
- Won't fit−843GB headroom
- Won't fit−875GB headroom
- Won't fit−715GB headroom
- Won't fit−715GB headroom
- Won't fit−459GB headroom
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −950GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −942GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −934GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −942GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −919GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −930GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −904GB | — |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −933GB | — |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Won't fit | −904GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −919GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −904GB | — |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Won't fit | −843GB | — |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Won't fit | −875GB | — |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Won't fit | −843GB | — |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Won't fit | −843GB | — |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Won't fit | −843GB | — |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Won't fit | −875GB | — |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Won't fit | −715GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Won't fit | −715GB | — |
| Mac Studio, M5 Ultra | 512GB | — | — | Won't fit | −459GB | — |
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 × 2.000 B/param (16 bits) + 0.254 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (35B active × 2.000 B/param) = 70.0 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.