~/best-gpu-for/gemma-4-26b
Google provider

Best GPU for Gemma 4 26B A4B

Ranked recommendations from the 42-GPU catalog for running Gemma 4 26B A4B (26B 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
11 GB weights · 0.97 GB headroom
Best quality
H100 80GB
80GB · runs at FP16/BF16
smallest card supporting the best available quant
Apple Silicon
M2 Max 64
64GB · runs at Q8_0
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 gemma-4-26b
GPUVRAMBest quantWeights at 15% safetyFit
nvidiaH100 80GB80GBFP16/BF16
52fitsappleM3 Max 9696GBFP16/BF16
52fitsnvidiaRTX Pro 600096GBFP16/BF16
52fitsappleM4 Max 128128GBFP16/BF16
52fitsappleM5 Max 128128GBFP16/BF16
52fitsnvidiaDGX Spark128GBFP16/BF16
52fitsintelGaudi 3128GBFP16/BF16
52fitsnvidiaH200141GBFP16/BF16
52fitsnvidia2× H100 NVL188GBFP16/BF16
52fitsappleM2 Ultra 192192GBFP16/BF16
52fitsnvidiaB200192GBFP16/BF16
52fitsamdMI300X192GBFP16/BF16
52fitsappleM3 Ultra 512512GBFP16/BF16
52fitsnvidiaRTX A600048GBQ8_0
28fitsnvidiaRTX 6000 Ada48GBQ8_0
28fitsnvidiaL40S48GBQ8_0
28fitsamdRadeon Pro W790048GBQ8_0
28fitsappleM2 Max 6464GBQ8_0
28fits

Next steps.

$ ./next
$ checkCan I run Gemma 4 26B A4B?see all quants →full quant matrix$ tuneTune the mathopen calculator →/calc?model=gemma-4-26b

Discussion.

$ gh discussion list

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