Llama-3.3-70B-Instruct 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 28 quants, from 15.9 GB (IQ1_S) to 141.1 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.
Llama‑3.3‑70B‑Instruct, a 70‑billion‑parameter instruction‑tuned LLM
Llama‑3.3‑70B‑Instruct is a 70‑billion‑parameter, instruction‑tuned language model from Meta. It supports multilingual dialogue in English, German, French, Italian, Portuguese, Hindi, Spanish, and Thai. The model is available in multiple GGUF quantizations for varied memory budgets.
Good at
- Multilingual instruction following
- Large context handling
- High‑quality text generation
Limits
- Requires large VRAM for full precision
- No vision or audio modalities
- License restricts commercial use
Best for
- Chatbot development
- Content creation
- Research prototyping
Who it’s for. Developers and researchers needing a large, instruction‑tuned model for multilingual dialogue. Suitable for teams with access to high‑memory GPUs or distributed inference setups.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 15.9 GB | 19.0 GB | Too big | Get file → | |
| IQ1_M | 17.1 GB | 20.3 GB | Too big | Get file → | |
| IQ2_XXS | 19.4 GB | 22.6 GB | Too big | Get file → | |
| IQ2_M | 24.3 GB | 27.5 GB | Too big | Get file → | |
| Q2_K | 26.4 GB | 29.6 GB | Too big | Get file → | |
| Q2_K_L | 26.6 GB | 29.8 GB | Too big | Get file → | |
| Q2_K_XL | 27.0 GB | 30.2 GB | Too big | Get file → | |
| IQ3_XXS | 27.7 GB | 30.8 GB | Too big | Get file → | |
| Q3_K_S | 30.9 GB | 34.1 GB | Too big | Get file → | |
| Q3_K_M | 34.3 GB | 37.5 GB | Too big | Get file → | |
| Q3_K_XL | 34.8 GB | 38.0 GB | Too big | Get file → | |
| IQ4_XS | 37.9 GB | 41.1 GB | Too big | Get file → | |
| IQ4_NL | 40.1 GB | 43.2 GB | Too big | Get file → | |
| Q4_0 | 40.1 GB | 43.3 GB | Too big | Get file → | |
| Q4_K_S | 40.3 GB | 43.5 GB | Too big | Get file → | |
| Q4_K_M | 42.5 GB | 45.7 GB | Too big | Get file → | |
| Q4_K_XL | 42.7 GB | 45.8 GB | Too big | Get file → | |
| Q4_1 | 44.3 GB | 47.5 GB | Too big | Get file → | |
| Q5_K_S | 48.7 GB | 51.8 GB | Too big | Get file → | |
| Q5_K_XL | 49.9 GB | 53.0 GB | Too big | Get file → | |
| Q5_K_M | 49.9 GB | 53.1 GB | Too big | Get file → | |
| Q6_K | 57.9 GB (2 parts) | 61.1 GB | Too big | Get file → | |
| Q6_K | 57.9 GB (2 parts) | 61.1 GB | Too big | Get file → | |
| Q6_K_XL | 61.2 GB (2 parts) | 64.3 GB | Too big | Get file → | |
| Q8_0 | 75.0 GB (2 parts) | 78.2 GB | Too big | Get file → | |
| Q8_K_XL | 81.2 GB (2 parts) | 84.4 GB | Too big | Get file → | |
| BF16 | 141.1 GB (3 parts) | 144.3 GB | Too big | Get file → | |
| F16 | 141.1 GB (4 parts) | 144.3 GB | Too big | Get file → |
For 8‑GB VRAM, use IQ1_S (15.85 GB file, ~19 GB VRAM). For 16‑GB VRAM, IQ1_M (17.15 GB, ~20.3 GB). For 24‑GB VRAM, Q2_K (26.38 GB, ~29.6 GB). Larger models need 32‑GB or more. Sharded Q6_K and Q8_0 split into two shards reduce peak memory slightly.
What GPU do you need for Llama-3.3-70B-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 (19.4 GB file, about 22.6 GB in use).
- 32 GB VRAM: the best quant that fits is IQ3_XXS (27.7 GB file, about 30.8 GB in use).
- 48 GB VRAM: the best quant that fits is Q4_1 (44.3 GB file, about 47.5 GB in use).
Which Llama-3.3-70B-Instruct quant should you choose?
Q4_K_M is the usual starting point for Llama-3.3-70B-Instruct: it balances file size and quality well. Here it is a 42.5 GB file.
The Q8_0 file is 75.0 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_S at 15.9 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 Llama-3.3-70B-Instruct
How much VRAM does Llama-3.3-70B-Instruct 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-3.3-70B-Instruct on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 19.0 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Llama-3.3-70B-Instruct GGUF?
It depends on the quant: IQ1_S is 15.9 GB, IQ1_M is 17.1 GB, IQ2_XXS is 19.4 GB, IQ2_M is 24.3 GB, Q2_K is 26.4 GB, Q2_K_L is 26.6 GB and 22 more.
What context length does Llama-3.3-70B-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 Llama-3.3-70B-Instruct GGUF?
The files are on Hugging Face in the unsloth/Llama-3.3-70B-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-3.3-70B-Instruct use?
Hugging Face lists the license as llama3.3. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run Llama‑3.3‑70B‑Instruct on 8 GB VRAM?
Use the IQ1_S quantization; it requires about 19 GB VRAM at 8k context, so 8 GB is insufficient.
Does this model support image input?
No, it is a text‑only model with no vision capabilities.
What license applies to this model?
The model is licensed under the llama3.3 license, which restricts commercial use.