~/best-gpu-for/qwen-3-6-27b
Alibaba provider

Best GPU for Qwen 3.6 27B

Ranked recommendations from the 42-GPU catalog for running Qwen 3.6 27B (27B 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
12 GB weights · 0.54 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 qwen-3-6-27b
GPUVRAMBest quantWeights at 15% safetyFit
nvidiaH100 80GB80GBFP16/BF16
54fitsappleM3 Max 9696GBFP16/BF16
54fitsnvidiaRTX Pro 600096GBFP16/BF16
54fitsappleM4 Max 128128GBFP16/BF16
54fitsappleM5 Max 128128GBFP16/BF16
54fitsnvidiaDGX Spark128GBFP16/BF16
54fitsintelGaudi 3128GBFP16/BF16
54fitsnvidiaH200141GBFP16/BF16
54fitsnvidia2× H100 NVL188GBFP16/BF16
54fitsappleM2 Ultra 192192GBFP16/BF16
54fitsnvidiaB200192GBFP16/BF16
54fitsamdMI300X192GBFP16/BF16
54fitsappleM3 Ultra 512512GBFP16/BF16
54fitsnvidiaRTX A600048GBQ8_0
29fitsnvidiaRTX 6000 Ada48GBQ8_0
29fitsnvidiaL40S48GBQ8_0
29fitsamdRadeon Pro W790048GBQ8_0
29fitsappleM2 Max 6464GBQ8_0
29fits

Next steps.

$ ./next
$ checkCan I run Qwen 3.6 27B?see all quants →full quant matrix$ tuneTune the mathopen calculator →/calc?model=qwen-3-6-27b

Discussion.

$ gh discussion list

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