Huihui GLM 4.7 Flash Abliterated — Hardware Requirements & GPU Compatibility
Chat
Huihui GLM 4.7 Flash Abliterated is a 31.2B-parameter open language model from huihui-ai in the GLM 4 family. It supports a context window of up to 202,752 tokens. At Q4_K_M it needs about 19.82 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 16 GB (Q2_K · 32K)Best on 24 GB (Q4_K_M · ~142 tok/s · 64K)Macs from 18 GB (Q2_K · ~34 tok/s)
MIT
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Huihui GLM 4.7 Flash Abliterated
One recommended way. Every other method, OS and quant is in the step-by-step guide.
Recommended
ollama run hf.co/mradermacher/Huihui-GLM-4.7-Flash-abliterated-i1-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 4.1 GB free · 16K context
Download: Huihui-GLM-4.7-Flash-abliterated.i1-Q4_K_M.gguf18.13 GBmradermacher
Other ways
each opens step-by-stepRun in cloud
Fits on RTX A5000 (24 GB) (4 GB headroom) · Q4_K_M
- Generation speed
- ~118 tok/s
- generation speed
- Cost per 1M output tokens
- $0.38
- per 1M output tokens
How Much VRAM Does Huihui GLM 4.7 Flash Abliterated Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 14.4 GB | 15.4 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~172 tok/s | |
| 16.3 GB | 17.3 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~160 tok/s | |
| 19.8 GB | 20.8 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 64K context · ~142 tok/s | |
| 23.3 GB | 24.3 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 16K context · ~128 tok/s | |
| 26.9 GB | 27.8 GB | 6-bit, very good — 6-bit quantization, very good quality | Does not fit: needs 27 GB | |
| 32.3 GB | 33.3 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Does not fit: needs 32.4 GB | |
| 63.5 GB | 64.5 GB | 16-bit · bfloat — Brain floating point 16 — preferred for training | Does not fit: needs 63.6 GB |
Selected Q4_K_M
4-bit medium quantization19.9GBat 16K context
- Weights
- 18.7 GB
- KV cache
- 0.9 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 4.1 GB free · ~142 tok/s
≥50 tok/s needs ≥258 GB/s · RTX 5090 ~196, RTX 3090 Ti ~142
Which Hardware Can Run Huihui GLM 4.7 Flash Abliterated?
Q4_K_M · 19.8 GBHuihui GLM 4.7 Flash Abliterated (Q4_K_M) requires 19.8 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 26+ GB is recommended. Using the full 203K context window can add up to 10.2 GB, bringing total usage to 30.0 GB. 8 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 122 devices with unified memory can run Huihui GLM 4.7 Flash Abliterated, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio M1 Max (32 GB).
20 GB
fits · 0.1 GB free · up to 16K ctx
1 GPU
RX 7900 XT~65Simulate output on RX 7900 XT32–40 GB
fits · 12.1 GB free · up to 198K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~196Simulate output on RTX 5090M5 Max36 GB~54Simulate output on Mac Studio M5 Max (36 GB)M532 GB~27Simulate output on MacBook Air 13" M5 (32 GB)Orin~41Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~26Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Alternatives to Huihui GLM 4.7 Flash Abliterated
GLM 4 family · 18 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| Hy Memory 30B A3BTencent | — | 19.1 GB | ~184 | same size |
| Llm Jp 4.1 32B A3b Thinkingllm-jp | — | 20.1 GB | ~188 | same size |
| Llm Jp 4.1 33B Thinkingllm-jp | — | 22.5 GB | ~35 | same size |
| Agnes 3.0 FlashAgnes-AI | — | 22.1 GB | ~35 | same size |
| Muse Glimmer 30BMeta | — | 19.6 GB | ~36 | same size |
| NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA | — | 19.3 GB | ~197 | same size |
Specifications
- Publisher
- huihui-ai
- Family
- GLM 4
- Parameters
- 31.2B
- Active parameters
- 4.6B per token
- Architecture
- Glm4MoeLiteForCausalLM
- Context Length
- 202,752 tokens
- Vocabulary Size
- 154,880
- Release Date
- 2026-01-22
- License
- MIT
Frequently Asked Questions
How much VRAM does Huihui GLM 4.7 Flash Abliterated need?
Huihui GLM 4.7 Flash Abliterated requires 19.8 GB of VRAM at Q4_K_M, or 63.5 GB at BF16. Full 203K context adds up to 10.2 GB (30.0 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 31.2B × 4.8 bits ÷ 8 = 18.7 GB
KV Cache + Overhead ≈ 1.1 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 11.3 GB (at full 203K context)
VRAM usage by quantization
Can NVIDIA GeForce RTX 4090 run Huihui GLM 4.7 Flash Abliterated?
Yes, at Q5_K_M (23.3 GB) or lower. Higher quantizations like Q6_K (26.9 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.
What's the best quantization for Huihui GLM 4.7 Flash Abliterated?
For Huihui GLM 4.7 Flash Abliterated, Q4_K_M (19.8 GB) offers the best balance of quality and VRAM usage. Q5_K_M (23.3 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 14.4 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Huihui GLM 4.7 Flash Abliterated on a Mac?
Yes — MacBook Pro 14" M3 Pro (18 GB) and 108 other Macs can run Huihui GLM 4.7 Flash Abliterated, the smallest at Q2_K. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 19.8 GB of usable unified memory (RAM minus macOS overhead), from Mac Mini M5 Pro (24 GB).
Can I run Huihui GLM 4.7 Flash Abliterated locally?
Yes — Huihui GLM 4.7 Flash Abliterated can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Huihui GLM 4.7 Flash Abliterated?
At Q4_K_M, Huihui GLM 4.7 Flash Abliterated can reach ~100 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~142 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 2.9 GB active ÷ (8000 × 0.65) = 0.56 ms, plus 47 layers × 0.055 ms = 2.58 ms, so 1000 ÷ 3.14 ms = ~318 tok/s
Estimated speed at Q4_K_M (19.8 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Huihui GLM 4.7 Flash Abliterated?
At Q4_K_M, the download is about 18.73 GB. The full-precision BF16 version is 62.44 GB. The smallest option (Q2_K) is 13.27 GB.
Which GPUs can run Huihui GLM 4.7 Flash Abliterated?
8 consumer GPUs can run Huihui GLM 4.7 Flash Abliterated at Q4_K_M. Top options include NVIDIA GeForce RTX 5090, AMD Radeon RX 7900 XT, AMD Radeon RX 7900 XTX. 1 GPU has plenty of headroom for comfortable inference.
Which devices can run Huihui GLM 4.7 Flash Abliterated?
122 devices with unified memory can run Huihui GLM 4.7 Flash Abliterated at Q4_K_M (19.8 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.