Can you run GLM-5.2 locally?
GLM family · Jun 2026 · 753.3B parameters · Hugging Face model card
GLM-5.2 is a mixture-of-experts model activating 8 of 256 experts per token released with open weights on Hugging Face: 753.33B total parameters, a 1,048,576-token context window, mit license. Running it locally needs roughly 512 GB of memory at Q4-class quantization (calculated) — multi-GPU server or large-unified-memory territory. Measured local speeds: not yet published here; we list unknowns as unknowns.
Minimum: 512 GB+ memory (Q4-class, calculated) · Recommended: 768 GB+ for comfortable context headroom (Q4-class, calculated)
How much VRAM does GLM-5.2 need at each quantization?
GLM-5.2 needs 497.2 GB of VRAM at Q4_K_M. The table below lists weights-only size and total VRAM including overhead for each common quantization level with 4k and 32k token contexts.
| Quantization | Bits / weight | Weights only | Total + KV @4k ctx | Total + KV @32k ctx |
|---|---|---|---|---|
| Q4_K_M | 4.8 | 452.0 GB | 497.2 GB | 497.2 GB |
| Q5_K_M | 5.7 | 536.8 GB | 590.4 GB | 590.4 GB |
| Q6_K | 6.6 | 621.5 GB | 683.7 GB | 683.7 GB |
| Q8_0 | 8.5 | 800.4 GB | 880.5 GB | 880.5 GB |
| FP16 | 16 | 1,506.7 GB | 1,657.3 GB | 1,657.3 GB |
Method: weights = parameters × bits-per-weight (Q4_K_M ≈ 4.8, Q5_K_M ≈ 5.7, Q6_K ≈ 6.6, Q8_0 ≈ 8.5, FP16 = 16), plus 10% loading overhead, plus KV cache from the verified architecture config (not applicable to this architecture). Source: HF-verified 2026-09-12 via api (repo safetensors + config.json); VRAM figures are documented calculations, not measurements
Which GPUs can run GLM-5.2 locally?
At Q4_K_M with a 4k context, GLM-5.2 needs 497.2 GB of VRAM. The lists below are computed live from our GPU database and grouped by how much headroom the card has. Disclosure: CompareAIHardware.com participates in the Amazon Associates program and earns from qualifying purchases through links on this page. Affiliate relationships do not influence our recommendations.
Needs 2+ GPUs to run GLM-5.2
One of these cards is too small on its own, but a pair (tensor or pipeline parallel, ~90% efficiency) covers the 497.2 GB requirement. See our multi-GPU guide for setup.
| GPU | VRAM | Bandwidth | Type | |
|---|---|---|---|---|
| AMD Instinct MI355X | 288 GB ×2 | 8000 GB/s | Data Center GPU | |
| NVIDIA B300 (Blackwell Ultra) | 288 GB ×2 | 8000 GB/s | Data Center GPU |