Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY needs about 38.0 GB of memory at Q4_K_M (a 35.8 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 11 quants, from 21.7 GB (Q2_K) to 62.9 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 6.6 GB.
Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY is a 59.2B-parameter GLM model in GGUF format, quantized by mradermacher. Its trained context length is 198K tokens. The repository has 11 quant files, from 21.7 GB (Q2_K) to 62.9 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 21.7 GB | 23.8 GB | Too big | Get file → | |
| Q3_K_S | 25.6 GB | 27.8 GB | Too big | Get file → | |
| Q3_K_M | 28.4 GB | 30.5 GB | Too big | Get file → | |
| Q3_K_L | 30.7 GB | 32.9 GB | Too big | Get file → | |
| IQ4_XS | 31.9 GB | 34.1 GB | Too big | Get file → | |
| Q4_K_S | 33.7 GB | 35.9 GB | Too big | Get file → | |
| Q4_K_M | 35.8 GB | 38.0 GB | Too big | Get file → | |
| Q5_K_S | 40.8 GB | 42.9 GB | Too big | Get file → | |
| Q5_K_M | 42.0 GB | 44.2 GB | Too big | Get file → | |
| Q6_K | 48.6 GB | 50.8 GB | Too big | Get file → | |
| Q8_0 | 62.9 GB | 65.1 GB | Too big | Get file → |
What GPU do you need for Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: the best quant that fits is Q2_K (21.7 GB file, about 23.8 GB in use).
- 32 GB VRAM: the best quant that fits is Q3_K_M (28.4 GB file, about 30.5 GB in use).
- 48 GB VRAM: the best quant that fits is Q5_K_M (42.0 GB file, about 44.2 GB in use).
Which Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY quant should you choose?
Q4_K_M is the usual starting point for Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY: it balances file size and quality well. Here it is a 35.8 GB file.
The Q8_0 file is 62.9 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is Q2_K at 21.7 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY
How much VRAM does Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY need?
About 38.0 GB at Q4_K_M with an 8K context window. That is the 35.8 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 23.8 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY GGUF?
It depends on the quant: Q2_K is 21.7 GB, Q3_K_S is 25.6 GB, Q3_K_M is 28.4 GB, Q3_K_L is 30.7 GB, IQ4_XS is 31.9 GB, Q4_K_S is 33.7 GB and 5 more.
What context length does Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY support?
The GGUF header lists a trained context length of 202,752 tokens (198K). The KV cache grows with the context you use: about 1.7 GB at 8K and 6.6 GB at 32K.
Where can I download Huihui GLM 4.7 Flash Abliterated 60B_DEPTHONLY GGUF?
The files are on Hugging Face in the mradermacher/Huihui-GLM-4.7-Flash-abliterated-60B_DEPTHONLY-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.