Models / GLM-5.2

Best home rig for GLM-5.2

Legacy model753.3B dense1 measured result model info ↗

Newer model available: GLM-5.3. GLM-5.2 is a legacy model; its numbers stay here for reference. source ↗

Real, measured generation speeds for GLM-5.2 on home hardware, fastest first. Different quants/engines aren't directly comparable.

Build	Quant	Engine	tok/s	Cost	Source
4x DGX Spark cluster running GLM-5.2 (unpruned)	QuantTrio Int4-Int8Mix	vLLM TP4 + MTP, single-stream mean	32.5 ⚡		github.com ↗

Rent before you buy: try this model on a cloud GPU for a few dollars first: Runpod (referral link; sign up with Google to get the new-user bonus)

Memory math: how much VRAM / RAM does GLM-5.2 need?

GLM-5.2 has 753.3B parameters (source: Hugging Face safetensors total: 753.3B parameters (MoE; the model card does not state the active count)). Weight size = parameters × bits-per-weight ÷ 8, before context (KV cache) and runtime overhead.

Format	Bits/weight	Weights	Smallest common memory size that holds the weights alone
Q8_0	8.50	800.5 GB	1024 GB
Q6_K	6.56	618.0 GB	1024 GB
Q5_K_M	5.70	537.1 GB	1024 GB
Q4_K_M	4.89	460.9 GB	512 GB
Q3_K_M	4.00	376.3 GB	512 GB
Q2_K	3.16	297.5 GB	512 GB

Bits-per-weight for GGUF quants are llama.cpp's own figures (llama.cpp quantize README). "As measured" rows apply the source's stated bits-per-weight to the parameter count; real files differ slightly because some layers stay at higher precision. Try other combinations in the VRAM calculator.

FAQ

How much VRAM does GLM-5.2 need?

The weights are about 460.9 GB at Q4_K_M (753.3B params × 4.89 bits ÷ 8, before context/KV cache). Add memory for context (KV cache) and the runtime. The smallest-memory setup with a measured result here, fully in memory, is 4x DGX Spark (GB10) (512 GB) at 32.5 tok/s.

What is the fastest measured hardware for GLM-5.2?

4x DGX Spark (GB10): 32.5 tok/s generation (vLLM TP4 + MTP, single-stream mean, QuantTrio Int4-Int8Mix), from https://github.com/tonyd2wild/GLM-5.2-QuantTrio-200K-4x-DGX-Spark.

Which GPUs have measured results for GLM-5.2?

4x DGX Spark (GB10).

Can it run on… 4x DGX Spark (GB10)