LL

llama 3.2b hf 32768 IQ4 NL

LlamaGGUF by matrixportalxUpdated 2024-09-01
Download on Hugging Face
Parameters3.18B
Context32K
Layers9
Architecturellama
Smallest quant1.8 GB
Quick answer

llama 3.2b hf 32768 IQ4 NL needs about 3.8 GB of memory at IQ4_NL (a 1.8 GB file) with an 8K context window, so it fits on a 4 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 6.0 GB.

llama 3.2b hf 32768 IQ4 NL is a 3.18B-parameter Llama model in GGUF format, quantized by matrixportalx. Its trained context length is 32K tokens. The repository has 1 quant file, from 1.8 GB (IQ4_NL) to 1.8 GB (IQ4_NL).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ4_NL1.8 GB 3.8 GB
Fits Get file →

What GPU do you need for llama 3.2b hf 32768 IQ4 NL?

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: the best quant that fits is IQ4_NL (1.8 GB file, about 3.8 GB in use).

Which llama 3.2b hf 32768 IQ4 NL quant should you choose?

IQ4_NL (1.8 GB) is the middle option in this repository. 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

Architecturellama
Parameters3.18B
Layers9
Trained context32,768 tokens
KV cache (fp16)180 KB per token

Source

Quantized bymatrixportalx
FamilyLlama
Downloads (HF)24
Last updated2024-09-01

Frequently asked questions about llama 3.2b hf 32768 IQ4 NL

How much VRAM does llama 3.2b hf 32768 IQ4 NL need?

About 3.8 GB at IQ4_NL with an 8K context window. That is the 1.8 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 llama 3.2b hf 32768 IQ4 NL on an 8 GB GPU?

Yes, with IQ4_NL: the 1.8 GB file needs about 3.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of llama 3.2b hf 32768 IQ4 NL GGUF?

It depends on the quant: IQ4_NL is 1.8 GB.

What context length does llama 3.2b hf 32768 IQ4 NL support?

The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.5 GB at 8K and 6.0 GB at 32K.

Where can I download llama 3.2b hf 32768 IQ4 NL GGUF?

The files are on Hugging Face in the matrixportalx/llama-3.2b-hf-32768-IQ4_NL-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

Similar models

Browse all →

More from matrixportalx

All by matrixportalx →

More Llama models

All Llama →

Other tiny models

All tiny models →