MoMo 72B Lora 1.8.7 DPO needs about 65.7 GB of memory at Q4_K_M (a 43.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 18 quants, from 13.3 GB (IQ4_XS) to 76.8 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 85.9 GB.
MoMo 72B Lora 1.8.7 DPO is a 72B-parameter Momo model in GGUF format, quantized by RichardErkhov. Its trained context length is 32K tokens. The repository has 18 quant files, from 13.3 GB (IQ4_XS) to 76.8 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ4_XS | 13.3 GB | 35.3 GB | Too big | Get file → | |
| Q4_0 | 14.3 GB | 36.3 GB | Too big | Get file → | |
| Q3_K_M | 24.3 GB | 46.3 GB | Too big | Get file → | |
| IQ4_NL | 24.5 GB | 46.4 GB | Too big | Get file → | |
| Q2_K | 27.1 GB | 49.1 GB | Too big | Get file → | |
| Q3_K_S | 31.6 GB | 53.5 GB | Too big | Get file → | |
| Q5_K | 32.3 GB | 54.3 GB | Too big | Get file → | |
| Q3_K | 35.3 GB | 57.3 GB | Too big | Get file → | |
| Q5_1 | 36.8 GB | 58.8 GB | Too big | Get file → | |
| Q3_K_L | 38.5 GB | 60.5 GB | Too big | Get file → | |
| Q4_K_S | 41.3 GB (2 parts) | 63.3 GB | Too big | Get file → | |
| Q4_K | 43.8 GB (2 parts) | 65.7 GB | Too big | Get file → | |
| Q4_K_M | 43.8 GB (2 parts) | 65.7 GB | Too big | Get file → | |
| Q4_1 | 45.4 GB (2 parts) | 67.4 GB | Too big | Get file → | |
| Q5_0 | 49.9 GB (2 parts) | 71.9 GB | Too big | Get file → | |
| Q5_K_S | 49.9 GB (2 parts) | 71.9 GB | Too big | Get file → | |
| Q6_K | 59.3 GB (2 parts) | 81.3 GB | Too big | Get file → | |
| Q8_0 | 76.8 GB (2 parts) | 98.8 GB | Too big | Get file → |
What GPU do you need for MoMo 72B Lora 1.8.7 DPO?
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: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: the best quant that fits is IQ4_NL (24.5 GB file, about 46.4 GB in use).
Which MoMo 72B Lora 1.8.7 DPO quant should you choose?
Q4_K_M is the usual starting point for MoMo 72B Lora 1.8.7 DPO: it balances file size and quality well. Here it is a 43.8 GB file.
The Q8_0 file is 76.8 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 IQ4_XS at 13.3 GB, with the quality trade-offs that come with fewer bits. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about MoMo 72B Lora 1.8.7 DPO
How much VRAM does MoMo 72B Lora 1.8.7 DPO need?
About 65.7 GB at Q4_K_M with an 8K context window. That is the 43.8 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 MoMo 72B Lora 1.8.7 DPO on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 35.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of MoMo 72B Lora 1.8.7 DPO GGUF?
It depends on the quant: IQ4_XS is 13.3 GB, Q4_0 is 14.3 GB, Q3_K_M is 24.3 GB, IQ4_NL is 24.5 GB, Q2_K is 27.1 GB, Q3_K_S is 31.6 GB and 12 more.
What context length does MoMo 72B Lora 1.8.7 DPO support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 21.5 GB at 8K and 85.9 GB at 32K.
Where can I download MoMo 72B Lora 1.8.7 DPO GGUF?
The files are on Hugging Face in the RichardErkhov/moreh_-_MoMo-72B-lora-1.8.7-DPO-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does MoMo 72B Lora 1.8.7 DPO use?
Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.