KQ

Kquant03 Raiden 16x3.43B

KquantGGUF by RichardErkhovUpdated 2024-08-31
Download on Hugging Face
Parameters35.8B
Context2K
Layers26
Architecturellama
Smallest quant20.2 GB
Quick answer

Kquant03 Raiden 16x3.43B needs about 30.1 GB of memory at Q4_K_M (a 26.8 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 22 quants, from 20.2 GB (IQ3_S) to 38.0 GB (Q8_0).

Kquant03 Raiden 16x3.43B is a 35.8B-parameter Kquant model in GGUF format, quantized by RichardErkhov. Its trained context length is 2K tokens. The repository has 22 quant files, from 20.2 GB (IQ3_S) to 38.0 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ3_S20.2 GB 21.4 GB
Too big Get file →
IQ3_XS20.2 GB 21.4 GB
Too big Get file →
Q2_K20.2 GB 21.4 GB
Too big Get file →
Q3_K_S20.2 GB 21.4 GB
Too big Get file →
Q4_020.2 GB 21.4 GB
Too big Get file →
IQ3_M20.4 GB 21.6 GB
Too big Get file →
IQ4_NL20.4 GB 21.6 GB
Too big Get file →
IQ4_XS20.4 GB 21.6 GB
Too big Get file →
Q3_K21.7 GB 22.9 GB
Too big Get file →
Q3_K_M21.7 GB 22.9 GB
Too big Get file →
Q4_122.4 GB 23.6 GB
Too big Get file →
Q3_K_L22.4 GB 23.6 GB
Too big Get file →
Q5_024.6 GB 25.8 GB
Too big Get file →
Q4_K_S24.7 GB 25.9 GB
Too big Get file →
Q4_K26.8 GB 28.0 GB
Too big Get file →
Q4_K_M26.8 GB 28.0 GB
Too big Get file →
Q5_126.9 GB 28.0 GB
Too big Get file →
Q5_K_S26.9 GB 28.0 GB
Too big Get file →
Q5_K28.7 GB 29.9 GB
Too big Get file →
Q5_K_M28.7 GB 29.9 GB
Too big Get file →
Q6_K38.0 GB 39.2 GB
Too big Get file →
Q8_038.0 GB 39.2 GB
Too big Get file →

What GPU do you need for Kquant03 Raiden 16x3.43B?

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 IQ4_NL (20.4 GB file, about 23.7 GB in use).
  • 32 GB VRAM: the best quant that fits is Q5_K (28.7 GB file, about 31.9 GB in use).
  • 48 GB VRAM: the best quant that fits is Q6_K (38.0 GB file, about 41.2 GB in use).

Which Kquant03 Raiden 16x3.43B quant should you choose?

Q4_K_M is the usual starting point for Kquant03 Raiden 16x3.43B: it balances file size and quality well. Here it is a 26.8 GB file.

The Q8_0 file is 38.0 GB, about 1.4 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 IQ3_S at 20.2 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

Architecturellama
Parameters35.8B (approximate, mixture of experts)
Layers26
Trained context2,048 tokens
KV cache (fp16)325 KB per token

Source

Quantized byRichardErkhov
FamilyKquant
Licenseapache-2.0
Downloads (HF)118
Last updated2024-08-31

Frequently asked questions about Kquant03 Raiden 16x3.43B

How much VRAM does Kquant03 Raiden 16x3.43B need?

About 30.1 GB at Q4_K_M with an 8K context window. That is the 26.8 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 Kquant03 Raiden 16x3.43B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 23.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Kquant03 Raiden 16x3.43B GGUF?

It depends on the quant: IQ3_S is 20.2 GB, IQ3_XS is 20.2 GB, Q2_K is 20.2 GB, Q3_K_S is 20.2 GB, Q4_0 is 20.2 GB, IQ3_M is 20.4 GB and 16 more.

What context length does Kquant03 Raiden 16x3.43B support?

The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 2.7 GB at 8K and 0.7 GB at 2K.

Where can I download Kquant03 Raiden 16x3.43B GGUF?

The files are on Hugging Face in the RichardErkhov/Kquant03_-_Raiden-16x3.43B-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 Kquant03 Raiden 16x3.43B 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 RichardErkhov

All by RichardErkhov →

Other large models

All large models →