LL

nyun-llama3-62B

LlamaGGUF by bartowskiUpdated 2024-06-03
Download on Hugging Face
Parameters62B
Context8K
Layers70
Architecturellama
Smallest quant14.8 GB
Quick answer

nyun-llama3-62B needs about 40.1 GB of memory at Q4_K_M (a 37.3 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 13 quants, from 14.8 GB (IQ1_M) to 124.0 GB (F16).

nyun-llama3-62B is a 62B-parameter Llama model in GGUF format, quantized by bartowski. Its trained context length is 8K tokens. The repository has 13 quant files, from 14.8 GB (IQ1_M) to 124.0 GB (F16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_M14.8 GB 17.6 GB
Too big Get file →
IQ2_XXS16.8 GB 19.7 GB
Too big Get file →
IQ2_M21.2 GB 24.1 GB
Too big Get file →
Q2_K23.2 GB 26.0 GB
Too big Get file →
IQ3_XXS24.1 GB 27.0 GB
Too big Get file →
Q3_K_S27.1 GB 29.9 GB
Too big Get file →
IQ3_M28.0 GB 30.9 GB
Too big Get file →
Q3_K_M30.1 GB 32.9 GB
Too big Get file →
IQ4_XS33.3 GB 36.2 GB
Too big Get file →
Q4_K_M37.3 GB 40.1 GB
Too big Get file →
Q5_K_M43.9 GB 46.7 GB
Too big Get file →
Q8_065.9 GB (2 parts) 68.7 GB
Too big Get file →
F16124.0 GB (3 parts) 126.9 GB
Too big Get file →

What GPU do you need for nyun-llama3-62B?

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 IQ2_XXS (16.8 GB file, about 19.7 GB in use).
  • 32 GB VRAM: the best quant that fits is IQ3_M (28.0 GB file, about 30.9 GB in use).
  • 48 GB VRAM: the best quant that fits is Q5_K_M (43.9 GB file, about 46.7 GB in use).

Which nyun-llama3-62B quant should you choose?

Q4_K_M is the usual starting point for nyun-llama3-62B: it balances file size and quality well. Here it is a 37.3 GB file.

The Q8_0 file is 65.9 GB, about 1.8 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 IQ1_M at 14.8 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

Architecturellama
Parameters62B
Layers70
Trained context8,192 tokens
KV cache (fp16)280 KB per token

Source

Quantized bybartowski
FamilyLlama
Downloads (HF)148
Last updated2024-06-03

Frequently asked questions about nyun-llama3-62B

How much VRAM does nyun-llama3-62B need?

About 40.1 GB at Q4_K_M with an 8K context window. That is the 37.3 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 nyun-llama3-62B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 17.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of nyun-llama3-62B GGUF?

It depends on the quant: IQ1_M is 14.8 GB, IQ2_XXS is 16.8 GB, IQ2_M is 21.2 GB, Q2_K is 23.2 GB, IQ3_XXS is 24.1 GB, Q3_K_S is 27.1 GB and 7 more.

What context length does nyun-llama3-62B support?

The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 2.3 GB at 8K and 2.3 GB at 8K.

Where can I download nyun-llama3-62B GGUF?

The files are on Hugging Face in the bartowski/nyun-llama3-62B-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 bartowski

All by bartowski →

More Llama models

All Llama →

Other large models

All large models →