Cheapest box to run Qwen3-Coder-Next at Q3_K_M
Alibaba’s Qwen3-Coder-Next is 80B 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-Coder-Next at Q3_K_M needs 41GB.
Cheapest Config that fits: Mac mini, M5 Pro, 64GB. Fits, 18GB headroom. Est. 105 tok/s.
Best offer today: $2,899 at Apple — in stock, checked 50 minutes ago.
Price history: 34 days recorded.
The Qwen3-Next hybrid again: 12 of 48 layers are full attention and the linear layers cache nothing per token.
Offers — Mac mini, M5 Pro, 64GB, 512GB
- Mac mini, M6, 16GB$1,099Won't fit−28GB headroom
- Mac mini, M6, 24GB$1,299Won't fit−20GB headroom
- Mac mini, M6, 32GB$1,499Won't fit−12GB headroom
- Mac mini, M5 Pro, 24GB$1,699Won't fit−20GB headroom
- Mac mini, M5 Pro, 48GB$2,499Tight4GB headroomest. 105 tok/s
- Mac Studio, M5 Max, 36GB$2,499Won't fit−8GB headroom
- Mac mini, M5 Pro, 64GB$2,899Fits18GB headroomest. 105 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostWon't fit−10GB headroom
- Fits18GB headroomest. 186 tok/s
- Mac Studio, M5 Max, 48GB$3,099Tight4GB headroomest. 209 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits18GB headroomest. 209 tok/s
- Fits80GB headroomest. 93.1 tok/s
- Fits48GB headroomest. 279 tok/s
- Fits79GB headroomest. 186 tok/s
- Fits80GB headroomest. 93.1 tok/s
- Fits79GB headroomest. 209 tok/s
- Fits48GB headroomest. 409 tok/s
- Fits207GB headroomest. 279 tok/s
- Fits207GB headroomest. 409 tok/s
- Fits463GB headroomest. 409 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −28GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Won't fit | −20GB | — |
| Mac mini, M6 | 32GB | $1,499 | $47 | Won't fit | −12GB | — |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Won't fit | −20GB | — |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Tight | 4GB | 105 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Won't fit | −8GB | — |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Fits | 18GB | 105 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Won't fit | −10GB | — |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Fits | 18GB | 186 tok/s |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Tight | 4GB | 209 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Fits | 18GB | 209 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Fits | 80GB | 93.1 tok/s |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Fits | 48GB | 279 tok/s |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Fits | 79GB | 186 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 80GB | 93.1 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 79GB | 209 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 48GB | 409 tok/s |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Fits | 207GB | 279 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 207GB | 409 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 463GB | 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 = 80B × 0.489 B/param (3.91 bits) + 0.025 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 M5 Pro 64GB runs
- Llama-4-Scout-17B-16EQ3_K_M, est. 18.5 tok/s
- Qwen3-Coder-NextQ5_K_M, est. 72.2 tok/s
- Llama-3.3-70BQ5_K_M, est. 4.0 tok/s
- Qwen3.5-35B-A3BQ8_0, est. 48.2 tok/s
- Qwen3-32BQ8_0, est. 5.9 tok/s
- Gemma-4-31BQ8_0, est. 6.1 tok/s
Best quant that still fits at 8K context, largest model first. Full list on the Config page.
Bandwidth 307GB/s, usable memory 59GB of 64GB.