← AI Hashrate

Best GPUs for GLM Models (4.5, 4.7, 5.x)

Our catalog tracks 5 GLM models (31.2B–753.9 parameters). For each, the table shows the cheapest retail GPU that fully fits it at Q4 / 8K context and the fastest retail GPU, with decode tok/s. Fit means weights + KV cache + 1 GB runtime overhead within 95% of VRAM — not weights alone. 3 of them exceed every retail GPU we track and are marked accordingly. Methodology.

#ModelParamsQ4/8K needs (GB)Best value GPU (fits)Fastest retail GPUBuy
1GLM-4.7-Flash31.2B MoE 3B act.18.76RX 7900 XT 20GB ≈169.7 tok/s est.RTX 5090 32GB ≈380.1 tok/s est.Search Amazon (affiliate)
2GLM-4.5-Air110.5B MoE 12B act.64.18Mac Studio M3 Ultra 96GB ≈43.4 tok/s est.Mac Studio M3 Ultra 96GB ≈43.4 tok/s est.Search Amazon (affiliate)
3GLM-4.5358.3B MoE 32B act.204.47No retail GPU fits
4GLM-5.2753.3B MoE 40B act.423.31No retail GPU fits
5GLM-5.1753.9B MoE 40B act.423.65No retail GPU fits

est. = bandwidth-based estimate; measured = curated real benchmark that overrides the estimate. GPU names link to the full fit check (4K/8K/32K fit table + VRAM breakdown). MSRP is the launch list price, not live retail — Amazon prices change; click a Buy link for the current price. Models marked 'No retail GPU fits' need more VRAM than any retail card in our catalog (up to 192 GB) — see the model page for datacenter-class options.

Popular fit checks

Compare the picks

Mac Studio M3 Ultra 96GB vs RTX 5090 32GB · Mac Studio M3 Ultra 96GB vs RX 7900 XT 20GB · RTX 5090 32GB vs RX 7900 XT 20GB

More lists: 7B-class models · 13B-class models · 30B-class models · 70B+ models · Under $500 · Under $1,000 · Under $2,000 · 8 GB VRAM · 12 GB VRAM · 16 GB VRAM · 24 GB VRAM · For Qwen3 · For DeepSeek · For Llama · For Gemma · For coding · For reasoning · MoE picks