Can I run GLM 5 locally?
GLM 5 by zai-org has 754B parameters (42B active per token, 256 experts, 8 used per token). At 4-bit with an 8K context it needs about 426 GB of GPU memory. It fits on 1 of 65 GPUs and Macs we track.
Memory needed by quant and context length
GB of GPU memory = weights + KV cache (FP16) + runtime overhead. Weights use real file sizes from unsloth/GLM-5-GGUF where available (marked •), otherwise parameter count × bits per weight. How we calculate.
| Quant | Quality | 4K ctx | 8K ctx | 32K ctx | 128K ctx |
|---|---|---|---|---|---|
| 2-bit (Q2_K) • | rough | 258 GB | 259 GB | 261 GB | 271 GB |
| 3-bit (Q3_K_M) • | usable | 337 GB | 337 GB | 340 GB | 350 GB |
| 4-bit (Q4_K_M) • | good, most popular | 426 GB | 426 GB | 429 GB | 439 GB |
| 5-bit (Q5_K_M) • | very good | 500 GB | 500 GB | 502 GB | 513 GB |
| 6-bit (Q6_K) • | near lossless | 578 GB | 578 GB | 581 GB | 591 GB |
| 8-bit (Q8_0) • | effectively lossless | 747 GB | 748 GB | 750 GB | 760 GB |
| 16-bit (F16/BF16) • | original weights | 1406 GB | 1406 GB | 1408 GB | 1419 GB |
Which GPUs and Macs run GLM 5?
Best quality quant that fits at 8K context. "Tight" = within 10% of usable memory. Speed is a ceiling from memory bandwidth (real speed is typically 50–80% of it).
Prefer an API?
Z.ai: GLM 5 is on OpenRouter at $1.00 per 1M input tokens and $3.20 per 1M output tokens. Price details and history →
Questions
How much VRAM does GLM 5 need?
About 426 GB at 4-bit (Q4_K_M) with an 8K context: 425 GB of weights (real GGUF file size), 0.8 GB of KV cache and 0.7 GB of runtime overhead. At 8-bit it needs about 748 GB.
What is the smallest GPU that runs GLM 5?
No single consumer GPU in our table fits it at Q4_K_M with 8K context.
Can I run GLM 5 on a Mac?
Not at Q4_K_M with 8K context on any Mac in our table, even a 512 GB Mac Studio, using the default GPU memory limit.
Model facts
| Hugging Face repo | zai-org/GLM-5 |
| Parameters | 753,864,139,008 (754B) · ~42B active (computed from config) |
| Layers / hidden size | 78 / 6144 |
| Attention | MLA; 64 heads, 64 KV heads, head dim 64 |
| KV cache per token (FP16) | 107.3 KiB |
| Max context | 202,752 tokens |
| License | mit |
| Released on HF | Feb 11, 2026 |
| Downloads (30d) / likes | 1,167,565 / 2,142 |
Sources: Hugging Face model API · config.json (zai-org/GLM-5) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.