Cheapest box to run Qwen3-Coder-480B-A35B-Instruct at Q8_0
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 Q8_0 needs 513GB.
Nothing on the Grid runs Qwen3-Coder-480B-A35B-Instruct at Q8_0 today. The largest Config, Mac Studio, M5 Ultra, 512GB, is 9GB short. At it fits with 266GB headroom.
- Mac mini, M6, 16GB$1,099Won't fit−500GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−492GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−484GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−492GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−469GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−480GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−454GB headroom
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−483GB headroom
- Won't fit−454GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−469GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−454GB headroom
- Won't fit−393GB headroom
- Won't fit−425GB headroom
- Won't fit−393GB headroom
- Won't fit−393GB headroom
- Won't fit−393GB headroom
- Won't fit−425GB headroom
- Won't fit−265GB headroom
- Won't fit−265GB headroom
- Won't fit−9GB headroom
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −500GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −492GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −484GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −492GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −469GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −480GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −454GB | — |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −483GB | — |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Won't fit | −454GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −469GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −454GB | — |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Won't fit | −393GB | — |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Won't fit | −425GB | — |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Won't fit | −393GB | — |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Won't fit | −393GB | — |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Won't fit | −393GB | — |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Won't fit | −425GB | — |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Won't fit | −265GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Won't fit | −265GB | — |
| Mac Studio, M5 Ultra | 512GB | — | — | Won't fit | −9GB | — |
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 × 1.063 B/param (8.5 bits) + 0.254 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (35B active × 1.063 B/param) = 37.2 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.