NU

Nusantara 7b Indo Chat

NusantaraGGUF by ukungUpdated 2024-05-20
Download on Hugging Face
Parameters7.72B
Context32K
Layers32
Architectureqwen2
Smallest quant3.1 GB
Quick answer

Nusantara 7b Indo Chat needs about 9.6 GB of memory at Q4_K_M (a 4.8 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 14 quants, from 3.1 GB (Q2_K) to 8.2 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 17.2 GB.

Nusantara 7b Indo Chat is a 7.72B-parameter Nusantara model in GGUF format, quantized by ukung. Its trained context length is 32K tokens. The repository has 14 quant files, from 3.1 GB (Q2_K) to 8.2 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K3.1 GB 7.9 GB
Tight Get file →
Q3_K_S3.6 GB 8.4 GB
Too big Get file →
Q3_K_M3.9 GB 8.7 GB
Too big Get file →
Q3_K_L4.2 GB 9.0 GB
Too big Get file →
Q4_04.5 GB 9.3 GB
Too big Get file →
Q4_K_S4.5 GB 9.3 GB
Too big Get file →
Q4_K_M4.8 GB 9.6 GB
Too big Get file →
Q4_15.0 GB 9.8 GB
Too big Get file →
Q5_05.4 GB 10.2 GB
Too big Get file →
Q5_K_S5.4 GB 10.2 GB
Too big Get file →
Q5_K_M5.5 GB 10.3 GB
Too big Get file →
Q5_15.8 GB 10.6 GB
Too big Get file →
Q6_K6.3 GB 11.1 GB
Too big Get file →
Q8_08.2 GB 13.0 GB
Too big Get file →

What GPU do you need for Nusantara 7b Indo Chat?

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: the best quant that fits is Q2_K (3.1 GB file, about 7.9 GB in use).
  • 12 GB VRAM: the best quant that fits is Q6_K (6.3 GB file, about 11.1 GB in use).
  • 16 GB VRAM: the best quant that fits is Q8_0 (8.2 GB file, about 13.0 GB in use).

Which Nusantara 7b Indo Chat quant should you choose?

Q4_K_M is the usual starting point for Nusantara 7b Indo Chat: it balances file size and quality well. Here it is a 4.8 GB file.

The Q8_0 file is 8.2 GB, about 1.7 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 3.1 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

Architectureqwen2
Parameters7.72B
Layers32
Trained context32,768 tokens
KV cache (fp16)512 KB per token

Source

Quantized byukung
FamilyNusantara
Licenseapache-2.0
Downloads (HF)63
Last updated2024-05-20

Frequently asked questions about Nusantara 7b Indo Chat

How much VRAM does Nusantara 7b Indo Chat need?

About 9.6 GB at Q4_K_M with an 8K context window. That is the 4.8 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 Nusantara 7b Indo Chat on an 8 GB GPU?

Yes, with Q2_K: the 3.1 GB file needs about 7.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Nusantara 7b Indo Chat GGUF?

It depends on the quant: Q2_K is 3.1 GB, Q3_K_S is 3.6 GB, Q3_K_M is 3.9 GB, Q3_K_L is 4.2 GB, Q4_0 is 4.5 GB, Q4_K_S is 4.5 GB and 8 more.

What context length does Nusantara 7b Indo Chat support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 4.3 GB at 8K and 17.2 GB at 32K.

Where can I download Nusantara 7b Indo Chat GGUF?

The files are on Hugging Face in the ukung/Nusantara-7b-Indo-Chat-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 Nusantara 7b Indo Chat use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Similar models

Browse all →

More from ukung

All by ukung →

More Nusantara models

All Nusantara →

Other small models

All small models →