LL

Kan LLaMa SFT v0.1

LlamaGGUF by TensoicUpdated 2024-04-06
Download on Hugging Face
Parameters6.88B
Context4K
Layers32
Architecturellama
Smallest quant4.2 GB
Quick answer

Kan LLaMa SFT v0.1 needs about 9.0 GB of memory at Q4_K_M (a 4.2 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 3 quants, from 4.2 GB (Q4_K_M) to 5.6 GB (Q6_K).

Kan LLaMa SFT v0.1 is a 6.88B-parameter Llama model in GGUF format, quantized by Tensoic. Its trained context length is 4K tokens. The repository has 3 quant files, from 4.2 GB (Q4_K_M) to 5.6 GB (Q6_K).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q4_K_M4.2 GB 6.8 GB
Fits Get file →
Q5_K_M4.9 GB 7.5 GB
Tight Get file →
Q6_K5.6 GB 8.3 GB
Too big Get file →

What GPU do you need for Kan LLaMa SFT v0.1?

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: the best quant that fits is Q6_K (5.6 GB file, about 10.4 GB in use).

Which Kan LLaMa SFT v0.1 quant should you choose?

Q4_K_M is the usual starting point for Kan LLaMa SFT v0.1: it balances file size and quality well. Here it is a 4.2 GB file.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters6.88B
Layers32
Trained context4,096 tokens
KV cache (fp16)512 KB per token

Source

Quantized byTensoic
FamilyLlama
Licenseapache-2.0
Downloads (HF)46
Last updated2024-04-06

Frequently asked questions about Kan LLaMa SFT v0.1

How much VRAM does Kan LLaMa SFT v0.1 need?

About 9.0 GB at Q4_K_M with an 8K context window. That is the 4.2 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 Kan LLaMa SFT v0.1 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 9.0 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Kan LLaMa SFT v0.1 GGUF?

It depends on the quant: Q4_K_M is 4.2 GB, Q5_K_M is 4.9 GB, Q6_K is 5.6 GB.

What context length does Kan LLaMa SFT v0.1 support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 4.3 GB at 8K and 2.1 GB at 4K.

Where can I download Kan LLaMa SFT v0.1 GGUF?

The files are on Hugging Face in the Tensoic/Kan-LLaMa-SFT-v0.1-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 Kan LLaMa SFT v0.1 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 Tensoic

All by Tensoic →

More Llama models

All Llama →

Other small models

All small models →