LL

Llama 3.1 70B Instruct Hf

LlamaGGUF by hierholzerUpdated 2024-12-11
Download on Hugging Face
Parameters70.6B
Context128K
Layers80
Architecturellama
Smallest quant30.9 GB
Quick answer

Llama 3.1 70B Instruct Hf needs about 45.7 GB of memory at Q4_K_M (a 42.5 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 7 quants, from 30.9 GB (Q3_K_S) to 49.9 GB (Q5_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.

Llama 3.1 70B Instruct Hf is a 70.6B-parameter Llama model in GGUF format, quantized by hierholzer. Its trained context length is 128K tokens. The repository has 7 quant files, from 30.9 GB (Q3_K_S) to 49.9 GB (Q5_K_M).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_S30.9 GB 34.1 GB
Too big Get file →
Q3_K_M34.3 GB 37.5 GB
Too big Get file →
Q3_K_L37.1 GB 40.3 GB
Too big Get file →
Q4_K_S40.3 GB 43.5 GB
Too big Get file →
Q4_K_M42.5 GB 45.7 GB
Too big Get file →
Q5_K_S48.7 GB 51.8 GB
Too big Get file →
Q5_K_M49.9 GB 53.1 GB
Too big Get file →

What GPU do you need for Llama 3.1 70B Instruct Hf?

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: the best quant that fits is Q4_K_M (42.5 GB file, about 45.7 GB in use).

Which Llama 3.1 70B Instruct Hf quant should you choose?

Q4_K_M is the usual starting point for Llama 3.1 70B Instruct Hf: it balances file size and quality well. Here it is a 42.5 GB file.

If memory is tight, the smallest file is Q3_K_S at 30.9 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
Parameters70.6B
Layers80
Trained context131,072 tokens
KV cache (fp16)320 KB per token

Source

Quantized byhierholzer
FamilyLlama
Licensemit
Downloads (HF)130
Last updated2024-12-11

Frequently asked questions about Llama 3.1 70B Instruct Hf

How much VRAM does Llama 3.1 70B Instruct Hf need?

About 45.7 GB at Q4_K_M with an 8K context window. That is the 42.5 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 3.1 70B Instruct Hf on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 34.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Llama 3.1 70B Instruct Hf GGUF?

It depends on the quant: Q3_K_S is 30.9 GB, Q3_K_M is 34.3 GB, Q3_K_L is 37.1 GB, Q4_K_S is 40.3 GB, Q4_K_M is 42.5 GB, Q5_K_S is 48.7 GB and 1 more.

What context length does Llama 3.1 70B Instruct Hf support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 2.7 GB at 8K and 10.7 GB at 32K.

Where can I download Llama 3.1 70B Instruct Hf GGUF?

The files are on Hugging Face in the hierholzer/Llama-3.1-70B-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 Llama 3.1 70B Instruct Hf 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.

Similar models

Browse all →

More from hierholzer

All by hierholzer →

More Llama models

All Llama →

Other large models

All large models →