Can I run GLM 4.6 locally?
GLM 4.6 by zai-org has 357B parameters (34B active per token, 160 experts, 8 used per token). At 4-bit with an 8K context it needs about 204 GB of GPU memory. It fits on 4 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-4.6-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 | 124 GB | 125 GB | 134 GB | 168 GB |
| 3-bit (Q3_K_M) • | usable | 161 GB | 162 GB | 171 GB | 206 GB |
| 4-bit (Q4_K_M) • | good, most popular | 203 GB | 204 GB | 213 GB | 247 GB |
| 5-bit (Q5_K_M) • | very good | 238 GB | 239 GB | 248 GB | 282 GB |
| 6-bit (Q6_K) • | near lossless | 275 GB | 277 GB | 285 GB | 320 GB |
| 8-bit (Q8_0) • | effectively lossless | 355 GB | 357 GB | 365 GB | 400 GB |
| 16-bit (F16/BF16) • | original weights | 667 GB | 668 GB | 677 GB | 711 GB |
Which GPUs and Macs run GLM 4.6?
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 4.6 is on OpenRouter at $0.50 per 1M input tokens and $2.00 per 1M output tokens. Price details and history →
Questions
How much VRAM does GLM 4.6 need?
About 204 GB at 4-bit (Q4_K_M) with an 8K context: 201 GB of weights (real GGUF file size), 2.9 GB of KV cache and 0.7 GB of runtime overhead. At 8-bit it needs about 357 GB.
What is the smallest GPU that runs GLM 4.6?
No single consumer GPU in our table fits it at Q4_K_M with 8K context; the smallest option that does is the Mac M3 Ultra 512GB (512 GB).
Can I run GLM 4.6 on a Mac?
Yes. At Q4_K_M with 8K context it fits on a Mac M3 Ultra 512GB or larger, using the default macOS GPU memory limit.
Model facts
| Hugging Face repo | zai-org/GLM-4.6 |
| Parameters | 356,785,898,816 (357B) · ~34B active (computed from config) |
| Layers / hidden size | 92 / 5120 |
| Attention | GQA; 96 heads, 8 KV heads, head dim 128 |
| KV cache per token (FP16) | 368.0 KiB |
| Max context | 202,752 tokens |
| License | mit |
| Released on HF | Sep 29, 2025 |
| Downloads (30d) / likes | 26,960 / 1,245 |
Sources: Hugging Face model API · config.json (zai-org/GLM-4.6) · GGUF files · GPU specs from vendor pages (list). Data fetched Nov 17, 2026. Spot an error? Report it.