GLM 4.6 Derestricted v3 — Hardware Requirements & GPU Compatibility
Chat
GLM 4.6 Derestricted v3 is a 356.8B-parameter open language model from ArliAI in the GLM 4 family. It supports a context window of up to 202,752 tokens. At Q4_K_M it needs about 214.69 GB of VRAM — see which GPUs and Macs can run it below.
1.0K downloads
73 likesSpecifications
- Publisher
- ArliAI
- Family
- GLM 4
- Parameters
- 356.8B
- Architecture
- Glm4MoeForCausalLM
- Context Length
- 202,752 tokens
- Vocabulary Size
- 151,552
- Release Date
- 2025-12-02
- License
- MIT
Get Started
HuggingFace
How Much VRAM Does GLM 4.6 Derestricted v3 Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| Q2_Kest. | 3.40 | 152.3 GB | 183.8 GB | 151.63 GB | 2-bit quantization with K-quant improvements |
| Q3_K_Mest. | 3.90 | 174.6 GB | 206.1 GB | 173.93 GB | 3-bit medium quantization |
| Q4_K_Mest. | 4.80 | 214.7 GB | 246.2 GB | 214.07 GB | 4-bit medium quantization — most popular sweet spot |
| Q5_K_Mest. | 5.70 | 254.8 GB | 286.3 GB | 254.21 GB | 5-bit medium quantization — good quality/size tradeoff |
| Q6_Kest. | 6.60 | 295.0 GB | 326.5 GB | 294.35 GB | 6-bit quantization, very good quality |
| Q8_0est. | 8.00 | 357.4 GB | 388.9 GB | 356.79 GB | 8-bit quantization, near-lossless |
| BF16est. | 16.00 | 714.2 GB | 745.7 GB | 713.57 GB | Brain floating point 16 — preferred for training |
est.= calculated VRAM estimate; no published GGUF file found for that quantization yet. Other rows are verified against real community uploads.
Which GPUs Can Run GLM 4.6 Derestricted v3?
Q4_K_M · 214.7 GBGLM 4.6 Derestricted v3 (Q4_K_M) requires 214.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 280+ GB is recommended. Using the full 203K context window can add up to 31.5 GB, bringing total usage to 246.2 GB. No single GPU has enough memory — multi-GPU or cluster setups are needed.
Which Devices Can Run GLM 4.6 Derestricted v3?
Q4_K_M · 214.7 GB3 devices with unified memory can run GLM 4.6 Derestricted v3, including NVIDIA DGX H100, NVIDIA DGX A100 640GB.
Runs great
— Plenty of headroomRelated Models
Frequently Asked Questions
- How much VRAM does GLM 4.6 Derestricted v3 need?
GLM 4.6 Derestricted v3 requires 214.7 GB of VRAM at Q4_K_M, or 714.2 GB at BF16. Full 203K context adds up to 31.5 GB (246.2 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 356.8B × 4.8 bits ÷ 8 = 214.1 GB
KV Cache + Overhead ≈ 0.6 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 32.1 GB (at full 203K context)
VRAM usage by quantization
Q4_K_M214.7 GBQ4_K_M + full context246.2 GB- Can NVIDIA GeForce RTX 5090 run GLM 4.6 Derestricted v3?
No — GLM 4.6 Derestricted v3 requires at least 152.3 GB at Q2_K, which exceeds the NVIDIA GeForce RTX 5090's 32 GB of VRAM.
- What's the best quantization for GLM 4.6 Derestricted v3?
For GLM 4.6 Derestricted v3, Q4_K_M (214.7 GB) offers the best balance of quality and VRAM usage. Q5_K_M (254.8 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 152.3 GB.
VRAM requirement by quantization
Q2_K152.3 GBQ4_K_M ★214.7 GBQ5_K_M254.8 GBQ6_K295.0 GBQ8_0357.4 GBBF16714.2 GB★ Recommended — best balance of quality and VRAM usage.
- Can I run GLM 4.6 Derestricted v3 on a Mac?
GLM 4.6 Derestricted v3 requires at least 152.3 GB at Q2_K, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run GLM 4.6 Derestricted v3 locally?
Yes — GLM 4.6 Derestricted v3 can run locally on consumer hardware. At Q4_K_M quantization it needs 214.7 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is GLM 4.6 Derestricted v3?
At Q4_K_M, GLM 4.6 Derestricted v3 can reach ~21 tok/s on AMD Instinct MI350X. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B300 → 8000 ÷ 214.7 × 0.65 = ~24 tok/s
Estimated speed at Q4_K_M (214.7 GB)
~24 tok/s~21 tok/s~21 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of GLM 4.6 Derestricted v3?
At Q4_K_M, the download is about 214.07 GB. The full-precision BF16 version is 713.57 GB. The smallest option (Q2_K) is 151.63 GB.
- Which GPUs can run GLM 4.6 Derestricted v3?
No single consumer GPU has enough VRAM to run GLM 4.6 Derestricted v3 at Q4_K_M (214.7 GB). Multi-GPU or professional hardware is required.
- Which devices can run GLM 4.6 Derestricted v3?
4 devices with unified memory can run GLM 4.6 Derestricted v3 at Q4_K_M (214.7 GB), including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), NVIDIA DGX A100 640GB, NVIDIA DGX H100. Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.