Cheapest box to run Qwen3-235B-A22B at Q8_0
Alibaba’s Qwen3-235B-A22B is 235B parameters, 22B 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.
Qwen3-235B-A22B at Q8_0 needs 252GB.
Cheapest Config that fits: Mac Studio, M5 Ultra, 512GB. Fits, 252GB headroom. Est. 25.7 tok/s.
No price yet: Ships late October 2026. Apple has not published a price.
Collecting price history — 1 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−239GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−231GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−223GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−231GB headroom
- Mac mini, M5 Pro, 48GB$2,499Won't fit−208GB headroom
- Mac Studio, M5 Max, 36GB$2,499Won't fit−219GB headroom
- Mac mini, M5 Pro, 64GB$2,899Won't fit−194GB headroom
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−222GB headroom
- Mac Studio, M5 Max, 48GB$3,099Won't fit−208GB headroom
- Mac Studio, M5 Max, 64GB$3,499Won't fit−194GB headroom
- Won't fit−132GB headroom
- Won't fit−132GB headroom
- Won't fit−132GB headroom
- Won't fit−164GB headroom
- Won't fit−4GB headroom
- Fits252GB headroomest. 25.7 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −239GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −231GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −223GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −231GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Won't fit | −208GB | — |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −219GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Won't fit | −194GB | — |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −222GB | — |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Won't fit | −208GB | — |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Won't fit | −194GB | — |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Won't fit | −132GB | — |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Won't fit | −132GB | — |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Won't fit | −132GB | — |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Won't fit | −164GB | — |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Won't fit | −4GB | — |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 252GB | 25.7 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 = 235B × 1.063 B/param (8.5 bits) + 0.193 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (22B active × 1.063 B/param) = 23.4 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
- DeepSeek-V3Q5_K_M, est. 22.9 tok/s
- Llama-4-Maverick-17B-128EQ8_0, est. 33.2 tok/s
- Qwen3-235B-A22BBF16, est. 13.6 tok/s
- Mistral-Large-2411BF16, est. 3.2 tok/s
- gpt-oss-120bBF16, est. 58.8 tok/s
- Llama-4-Scout-17B-16EBF16, 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.