QW

Undi95_MG-FinalMix-72B

QwenGGUF by Undi95Updated 2024-06-26
Download on Hugging Face
Parameters72.7B
Context32K
Layers80
Architectureqwen2
Smallest quant37.7 GB
Quick answer

Undi95_MG-FinalMix-72B needs about 50.6 GB of memory at Q4_K_M (a 47.4 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 2 quants, from 37.7 GB (Q3_K_M) to 47.4 GB (Q4_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.

Undi95_MG-FinalMix-72B is a 72.7B-parameter Qwen model in GGUF format, quantized by Undi95. Its trained context length is 32K tokens. The repository has 2 quant files, from 37.7 GB (Q3_K_M) to 47.4 GB (Q4_K_M).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_M37.7 GB 40.9 GB
Too big Get file →
Q4_K_M47.4 GB 50.6 GB
Too big Get file →

What GPU do you need for Undi95_MG-FinalMix-72B?

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 Q3_K_M (37.7 GB file, about 40.9 GB in use).

Which Undi95_MG-FinalMix-72B quant should you choose?

Q4_K_M is the usual starting point for Undi95_MG-FinalMix-72B: it balances file size and quality well. Here it is a 47.4 GB file.

If memory is tight, the smallest file is Q3_K_M at 37.7 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architectureqwen2
Parameters72.7B
Layers80
Trained context32,768 tokens
KV cache (fp16)320 KB per token

Source

Quantized byUndi95
FamilyQwen
Downloads (HF)11
Last updated2024-06-26

Frequently asked questions about Undi95_MG-FinalMix-72B

How much VRAM does Undi95_MG-FinalMix-72B need?

About 50.6 GB at Q4_K_M with an 8K context window. That is the 47.4 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 Undi95_MG-FinalMix-72B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 40.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Undi95_MG-FinalMix-72B GGUF?

It depends on the quant: Q3_K_M is 37.7 GB, Q4_K_M is 47.4 GB.

What context length does Undi95_MG-FinalMix-72B support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.

Where can I download Undi95_MG-FinalMix-72B GGUF?

The files are on Hugging Face in the Undi95/MG-FinalMix-72B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Other GGUFs of Qwen2-72B-Instruct

Similar models

Browse all →

More from Undi95

All by Undi95 →

More Qwen models

All Qwen →

Other large models

All large models →