Cheapest box to run Qwen3.5-35B-A3B at Q3_K_M
Alibaba’s Qwen3.5-35B-A3B is 35B parameters, 3B 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.
Qwen3.5-35B-A3B at Q3_K_M needs 18GB.
Cheapest Config that fits: Mac mini, M6, 32GB. Fits, 11GB headroom. Est. 58.0 tok/s.
Best offer today: $1,499 at Apple — in stock, checked 32 minutes ago.
Price history: 34 days recorded.
A hybrid: only 10 of 40 layers are full attention. The Gated DeltaNet layers keep a constant-size state and cache nothing per token, so the K/V figure counts the full-attention layers alone.
Offers — Mac mini, M6, 32GB, 512GB
- Mac mini, M6, 16GB$1,099Won't fit−5GB headroom
- Mac mini, M6, 24GB$1,299Tight3GB headroomest. 58.0 tok/s
- Mac mini, M6, 32GB$1,499Fits11GB headroomest. 58.0 tok/s
- Mac mini, M5 Pro, 24GB$1,699Tight3GB headroomest. 105 tok/s
- Mac mini, M5 Pro, 48GB$2,499Fits26GB headroomest. 105 tok/s
- Mac Studio, M5 Max, 36GB$2,499Fits15GB headroomest. 157 tok/s
- Mac mini, M5 Pro, 64GB$2,899Fits40GB headroomest. 105 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostFits12GB headroomest. 611 tok/s
- Fits40GB headroomest. 186 tok/s
- Mac Studio, M5 Max, 48GB$3,099Fits26GB headroomest. 209 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits40GB headroomest. 209 tok/s
- Fits102GB headroomest. 93.1 tok/s
- Fits70GB headroomest. 279 tok/s
- Fits102GB headroomest. 186 tok/s
- Fits102GB headroomest. 93.1 tok/s
- Fits102GB headroomest. 209 tok/s
- Fits70GB headroomest. 409 tok/s
- Fits230GB headroomest. 279 tok/s
- Fits230GB headroomest. 409 tok/s
- Fits486GB headroomest. 409 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −5GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Tight | 3GB | 58.0 tok/s |
| Mac mini, M6 | 32GB | $1,499 | $47 | Fits | 11GB | 58.0 tok/s |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Tight | 3GB | 105 tok/s |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Fits | 26GB | 105 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Fits | 15GB | 157 tok/s |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Fits | 40GB | 105 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Fits | 12GB | 611 tok/s |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Fits | 40GB | 186 tok/s |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Fits | 26GB | 209 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Fits | 40GB | 209 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Fits | 102GB | 93.1 tok/s |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Fits | 70GB | 279 tok/s |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Fits | 102GB | 186 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 102GB | 93.1 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 102GB | 209 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 70GB | 409 tok/s |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Fits | 230GB | 279 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 230GB | 409 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 486GB | 409 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 = 35B × 0.489 B/param (3.91 bits) + 0.021 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (3B active × 0.489 B/param) = 1.5 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 mini M6 32GB runs
- Qwen3.5-35B-A3BQ5_K_M, est. 40.0 tok/s
- Qwen3-32BQ5_K_M, est. 4.9 tok/s
- Gemma-4-31BQ6_K, est. 4.4 tok/s
- Qwen3-30B-A3BQ6_K, est. 31.4 tok/s
- Gemma-3-27BQ6_K, est. 5.0 tok/s
- Qwen3.6-27BQ6_K, est. 5.0 tok/s
Best quant that still fits at 8K context, largest model first. Full list on the Config page.
Bandwidth 170GB/s, usable memory 29GB of 32GB.