LL

RYS Llama 3 Huge Instruct

LlamaGGUF by belisariusUpdated 2024-11-21
Download on Hugging Face
Parameters99.6B
Context8K
Layers114
Architecturellama
Smallest quant48.1 GB
Quick answer

RYS Llama 3 Huge Instruct needs about 61.1 GB of memory at Q4_K_S (a 56.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 3 quants, from 48.1 GB (Q3_K_M) to 199.3 GB (BF16).

RYS Llama 3 Huge Instruct is a 99.6B-parameter Llama model in GGUF format, quantized by belisarius. Its trained context length is 8K tokens. The repository has 3 quant files, from 48.1 GB (Q3_K_M) to 199.3 GB (BF16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_M48.1 GB (2 parts) 52.5 GB
Too big Get file →
Q4_K_S56.8 GB (3 parts) 61.1 GB
Too big Get file →
BF16199.3 GB (5 parts) 203.6 GB
Too big Get file →

What GPU do you need for RYS Llama 3 Huge Instruct?

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: no quant fits at an 8K context.

Which RYS Llama 3 Huge Instruct quant should you choose?

Q4_K_S (56.8 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.

If memory is tight, the smallest file is Q3_K_M at 48.1 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
Parameters99.6B
Layers114
Trained context8,192 tokens
KV cache (fp16)456 KB per token

Source

Quantized bybelisarius
FamilyLlama
Licensemit
Downloads (HF)9
Last updated2024-11-21

Frequently asked questions about RYS Llama 3 Huge Instruct

How much VRAM does RYS Llama 3 Huge Instruct need?

About 61.1 GB at Q4_K_S with an 8K context window. That is the 56.8 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 RYS Llama 3 Huge Instruct on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 52.5 GB, so you would need a larger GPU or CPU offloading.

What is the file size of RYS Llama 3 Huge Instruct GGUF?

It depends on the quant: Q3_K_M is 48.1 GB, Q4_K_S is 56.8 GB, BF16 is 199.3 GB.

What context length does RYS Llama 3 Huge Instruct support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 3.8 GB at 8K and 3.8 GB at 8K.

Where can I download RYS Llama 3 Huge Instruct GGUF?

The files are on Hugging Face in the belisarius/RYS-Llama-3-Huge-Instruct-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 RYS Llama 3 Huge Instruct use?

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

Other GGUFs of RYS-Llama-3-Huge-Instruct

Similar models

Browse all →

More from belisarius

All by belisarius →

More Llama models

All Llama →

Other large models

All large models →