Cheapest box to run Devstral-Small-2507 at Q3_K_M
Mistral’s Devstral-Small-2507 is 24B parameters, dense. 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.
Devstral-Small-2507 at Q3_K_M needs 14GB.
Cheapest Config that fits: Mac mini, M6, 24GB. Fits, 7GB headroom. Est. 9.4 tok/s.
Best offer today: $1,299 at Expercom — in stock, checked 34 hours ago.
Collecting price history — 3 of 90 days. No timing verdict yet.
Offers — Mac mini, M6, 24GB, 512GB
- Mac mini, M6, 16GB$1,099Won't fit−1GB headroom
- Mac mini, M6, 24GB$1,299Fits7GB headroomest. 9.4 tok/s
- Mac mini, M6, 32GB$1,499Fits15GB headroomest. 9.4 tok/s
- Mac mini, M5 Pro, 24GB$1,699Fits7GB headroomest. 17.0 tok/s
- Mac mini, M5 Pro, 48GB$2,499Fits30GB headroomest. 17.0 tok/s
- Mac Studio, M5 Max, 36GB$2,499Fits19GB headroomest. 34.0 tok/s
- Mac mini, M5 Pro, 64GB$2,899Fits45GB headroomest. 17.0 tok/s
- GeForce RTX 5090, 32GB$2,999.99+hostFits16GB headroomest. 99.3 tok/s
- Fits45GB headroomest. 30.3 tok/s
- Mac Studio, M5 Max, 48GB$3,099Fits30GB headroomest. 34.0 tok/s
- Mac Studio, M5 Max, 64GB$3,499Fits45GB headroomest. 34.0 tok/s
- Fits106GB headroomest. 15.1 tok/s
- Fits74GB headroomest. 45.4 tok/s
- Fits106GB headroomest. 30.3 tok/s
- Fits106GB headroomest. 15.1 tok/s
- Fits106GB headroomest. 34.0 tok/s
- Fits74GB headroomest. 66.5 tok/s
- Fits234GB headroomest. 45.4 tok/s
- Fits234GB headroomest. 66.5 tok/s
- Fits490GB headroomest. 66.5 tok/s
| Config | Memory | Price | $/GB | Fit | Headroom | Est. speed |
|---|---|---|---|---|---|---|
| Mac mini, M6 | 16GB | $1,099 | $69 | Won't fit | −1GB | — |
| Mac mini, M6 | 24GB | $1,299 | $54 | Fits | 7GB | 9.4 tok/s |
| Mac mini, M6 | 32GB | $1,499 | $47 | Fits | 15GB | 9.4 tok/s |
| Mac mini, M5 Pro | 24GB | $1,699 | $71 | Fits | 7GB | 17.0 tok/s |
| Mac mini, M5 Pro | 48GB | $2,499 | $52 | Fits | 30GB | 17.0 tok/s |
| Mac Studio, M5 Max | 36GB | $2,499 | $69 | Fits | 19GB | 34.0 tok/s |
| Mac mini, M5 Pro | 64GB | $2,899 | $45 | Fits | 45GB | 17.0 tok/s |
| GeForce RTX 5090 | 32GB | $2,999.99+host | $94 | Fits | 16GB | 99.3 tok/s |
| Mac Studio, M4 Max, refurb | 64GB | $3,049 | $48 | Fits | 45GB | 30.3 tok/s |
| Mac Studio, M5 Max | 48GB | $3,099 | $65 | Fits | 30GB | 34.0 tok/s |
| Mac Studio, M5 Max | 64GB | $3,499 | $55 | Fits | 45GB | 34.0 tok/s |
| Ascent GX10, GB10 Grace Blackwell | 128GB | $3,999.99 | $31 | Fits | 106GB | 15.1 tok/s |
| Mac Studio, M3 Ultra, refurb | 96GB | $4,329 | $45 | Fits | 74GB | 45.4 tok/s |
| Mac Studio, M4 Max, refurb | 128GB | $4,409 | $34 | Fits | 106GB | 30.3 tok/s |
| DGX Spark, GB10 Grace Blackwell | 128GB | $4,699 | $37 | Fits | 106GB | 15.1 tok/s |
| Mac Studio, M5 Max | 128GB | $5,099 | $40 | Fits | 106GB | 34.0 tok/s |
| Mac Studio, M5 Ultra | 96GB | $5,499 | $57 | Fits | 74GB | 66.5 tok/s |
| Mac Studio, M3 Ultra, refurb | 256GB | $7,729 | $30 | Fits | 234GB | 45.4 tok/s |
| Mac Studio, M5 Ultra | 256GB | $9,499 | $37 | Fits | 234GB | 66.5 tok/s |
| Mac Studio, M5 Ultra | 512GB | — | — | Fits | 490GB | 66.5 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 = 24B × 0.489 B/param (3.91 bits) + 0.164 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.65 ÷ (24B active × 0.489 B/param) = 11.7 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 24GB runs
- Qwen3-32BQ3_K_M, est. 7.1 tok/s
- Qwen3-30B-A3BQ4_K_M, est. 42.7 tok/s
- Gemma-3-27BQ4_K_M, est. 6.8 tok/s
- Devstral-Small-2507Q5_K_M, est. 6.5 tok/s
- gpt-oss-20bQ6_K, est. 28.8 tok/s
- Phi-4Q8_0, est. 7.4 tok/s
Best quant that still fits at 8K context, largest model first. Full list on the Config page.
Bandwidth 170GB/s, usable memory 21GB of 24GB.