~/best-gpu-for/mistral-small-3
Mistral provider

Best GPU for Mistral Small 3

Ranked recommendations from the 42-GPU catalog for running Mistral Small 3 (24B params). Top picks by quality, then by cost.

Top picks.

$ ./rank --by quality
Tightest budget
RTX 4060 Ti 16GB
16GB · runs at Q3_K_M
10 GB weights · 1.8 GB headroom
Best quality
M2 Max 64
64GB · runs at FP16/BF16
smallest card supporting the best available quant
Apple Silicon
M2 Max 64
64GB · runs at FP16/BF16
smallest M-series that fits
Datacenter (FP16 ambition)
H100 80GB
80GB · runs at FP16/BF16
best-quality datacenter card that fits

Full ranked list.

$ ./vrambudget --rank mistral-small-3
GPUVRAMBest quantWeights at 15% safetyFit
appleM2 Max 6464GBFP16/BF16
48fitsappleM3 Max 6464GBFP16/BF16
48fitsappleM4 Pro 6464GBFP16/BF16
48fitsappleM5 Pro 6464GBFP16/BF16
48fitsnvidiaH100 80GB80GBFP16/BF16
48fitsappleM3 Max 9696GBFP16/BF16
48fitsnvidiaRTX Pro 600096GBFP16/BF16
48fitsappleM4 Max 128128GBFP16/BF16
48fitsappleM5 Max 128128GBFP16/BF16
48fitsnvidiaDGX Spark128GBFP16/BF16
48fitsintelGaudi 3128GBFP16/BF16
48fitsnvidiaH200141GBFP16/BF16
48fitsnvidia2× H100 NVL188GBFP16/BF16
48fitsappleM2 Ultra 192192GBFP16/BF16
48fitsnvidiaB200192GBFP16/BF16
48fitsamdMI300X192GBFP16/BF16
48fitsappleM3 Ultra 512512GBFP16/BF16
48fitsnvidiaRTX A600048GBQ8_0
26fits

Next steps.

$ ./next
$ checkCan I run Mistral Small 3?see all quants →full quant matrix$ tuneTune the mathopen calculator →/calc?model=mistral-small-3

Discussion.

$ gh discussion list

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