GLM-4.6V
GLM-4.6V is an open-weights language model from Z.ai, released December 8, 2025. It has 108B parameters (mixture-of-experts, 14B active). The context window is 125K tokens. API price: $0.3 input / $0.9 output per million tokens.
Check it against your own PC →
Can you run GLM-4.6V locally?
The smallest common GPU it runs on is 8 GB of VRAM (with 32 GB of RAM): IQ2_XXS, GPU + experts in RAM, roughly 15 tokens per second. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Too large | – | – | – | – |
| 8 GB VRAM | Runs · GPU + CPU experts | IQ2_XXS (heavy loss) | 7.2 GB | 24.2 GB | ~15 |
| 12 GB VRAM | Runs · GPU + CPU experts | IQ2_XXS (heavy loss) | 11.2 GB | 20.2 GB | ~18 |
| 16 GB VRAM | Runs · GPU + CPU experts | IQ2_M (heavy loss) | 15.2 GB | 24.8 GB | ~15 |
| 24 GB VRAM | Runs · GPU + CPU experts | IQ3_XXS (noticeable loss) | 23.2 GB | 21.7 GB | ~18 |
| 32 GB VRAM | Runs · GPU + CPU experts | Q3_K_M (okay) | 31.2 GB | 25.1 GB | ~17 |
Ready-made quantizations on Hugging Face: unsloth/GLM-4.6V-GGUF · ggml-org/GLM-4.6V-GGUF · mradermacher/GLM-4.6V-i1-GGUF · bartowski/zai-org_GLM-4.6V-GGUF · mradermacher/GLM-4.6V-GGUF · steampunque/GLM-4.6V-MP-GGUF
Benchmark results
Epoch AI has not published results for GLM-4.6V yet. New models are usually evaluated within days to weeks.
Links
Hugging Face: zai-org/GLM-4.6V
More from Z.ai
GLM-5.3-FlashX · GLM 5.3 Flash BF16 · GLM 5.3 BF16 · GLM-5.3-Flash · GLM-5.3 · GLM-5.2 · GLM-5.1 · GLM-5V-Turbo