~/can-i-run/mistral-small-3
Mistral provider

Can I run Mistral Small 3?

Short answer: yes, on a M2 Max 64 (64GB) at FP16/BF16. Long answer below.

mistral-small:24bLM StudiovLLMMLXoMLX

The math, in one paragraph.

$ ./vrambudget --explain mistral-small-3

Mistral Small 3 has 24B parameters. At FP16 that's 48 GB of raw weights. Quantization shrinks that, but you also need budget for the KV cache (definition), framework overhead, and safety headroom. The rule of thumb: real usable budget on a card is roughly its nameplate VRAM minus 25%. That's how the table below was computed.

What hardware actually fits.

$ grep "fits" gpus.json
FP16/BF16
48GB
17 GPUs fit
M2 Max 6464GBM3 Max 6464GBM4 Pro 6464GBM5 Pro 6464GB+ 13 more
Q8_0
26GB
21 GPUs fit
RTX A600048GBRTX 6000 Ada48GBL40S48GBRadeon Pro W790048GB+ 17 more
Q5_K_M
17GB
27 GPUs fit
RTX 309024GBRTX 3090 Ti24GBRTX 409024GBRX 7900 XTX24GB+ 23 more
Q4_K_M
14GB
27 GPUs fit
RTX 309024GBRTX 3090 Ti24GBRTX 409024GBRX 7900 XTX24GB+ 23 more
Q3_K_M
10GB
33 GPUs fit
RTX 4060 Ti 16GB16GBRTX 4070 Ti Super16GBRTX 408016GBRTX 4080 Super16GB+ 29 more

Pick your path.

$ ls strategies/
Tightest budget

Smallest GPU that fits Mistral Small 3 at any quant: M2 Max 64 at FP16/BF16.

Reference quality (FP16)

Lossless inference needs 48 GB. Pick from 17 cards.

Best quality on a 24GB card

Q5_K_M fits comfortably (17 GB weights).

Tune the math yourself

Open the calculator pre-tuned for Mistral Small 3: ↗ /calc?model=mistral-small-3

See the full model page.

$ ./open
$ openMistral Small 324BMistral · ctx 32K$ rankBest GPU for Mistral Small 3see the ranking →ranked by fit quality

Discussion.

$ gh discussion list

// sign in with github to leave a comment. threads live in the repo's discussions tab.