Meta Llama 3.1 70B Instruct needs about 44.4 GB of memory at Q4_0 (a 41.2 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 12 quants, from 18.1 GB (IQ1_M) to 141.1 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.
Meta Llama 3.1 70B Instruct is a 70.6B-parameter Llama model in GGUF format, quantized by xaskasdf. Its trained context length is 128K tokens. The repository has 12 quant files, from 18.1 GB (IQ1_M) to 141.1 GB (F16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_M | 18.1 GB | 21.3 GB | Too big | Get file → | |
| IQ2_XXS | 20.5 GB | 23.7 GB | Too big | Get file → | |
| IQ2_S | 23.6 GB | 26.8 GB | Too big | Get file → | |
| Q2_K | 27.6 GB | 30.8 GB | Too big | Get file → | |
| Q3_K | 35.5 GB | 38.7 GB | Too big | Get file → | |
| Q4_0 | 41.2 GB | 44.4 GB | Too big | Get file → | |
| Q4_K | 43.8 GB | 46.9 GB | Too big | Get file → | |
| Q5_0 | 49.9 GB | 53.1 GB | Too big | Get file → | |
| Q5_K | 51.2 GB (2 parts) | 54.4 GB | Too big | Get file → | |
| Q6_K | 59.1 GB (2 parts) | 62.3 GB | Too big | Get file → | |
| Q8_0 | 76.0 GB (2 parts) | 79.1 GB | Too big | Get file → | |
| F16 | 141.1 GB (3 parts) | 144.3 GB | Too big | Get file → |
What GPU do you need for Meta Llama 3.1 70B 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: the best quant that fits is IQ2_XXS (20.5 GB file, about 23.7 GB in use).
- 32 GB VRAM: the best quant that fits is Q2_K (27.6 GB file, about 30.8 GB in use).
- 48 GB VRAM: the best quant that fits is Q4_K (43.8 GB file, about 46.9 GB in use).
Which Meta Llama 3.1 70B Instruct quant should you choose?
Q4_0 (41.2 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 IQ1_M at 18.1 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
Source
Frequently asked questions about Meta Llama 3.1 70B Instruct
How much VRAM does Meta Llama 3.1 70B Instruct need?
About 44.4 GB at Q4_0 with an 8K context window. That is the 41.2 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 Meta Llama 3.1 70B Instruct on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 21.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Meta Llama 3.1 70B Instruct GGUF?
It depends on the quant: IQ1_M is 18.1 GB, IQ2_XXS is 20.5 GB, IQ2_S is 23.6 GB, Q2_K is 27.6 GB, Q3_K is 35.5 GB, Q4_0 is 41.2 GB and 6 more.
What context length does Meta Llama 3.1 70B Instruct 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 Meta Llama 3.1 70B Instruct GGUF?
The files are on Hugging Face in the xaskasdf/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 Meta Llama 3.1 70B Instruct use?
Hugging Face lists the license as llama3.1. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.