LL

Llama-3.3-70B-Instruct

LlamaGGUF by MungertUpdated 2025-09-24
Download on Hugging Face
Parameters70.6B
Context128K
Layers80
Architecturellama
Smallest quant18.2 GB
Quick answer

Llama-3.3-70B-Instruct needs about 46.0 GB of memory at Q4_K_M (a 42.8 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 33 quants, from 18.2 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, 70.6B parameters, 131072‑token context

Llama‑3.3‑70B‑Instruct is a 70.6B‑parameter instruction‑tuned model with a 131072‑token context window. It supports tool usage and is available in multiple quantizations from IQ‑1 to Q8‑0. The model is released under the llama3.3 license.

Good at

  • Tool‑aware instruction following
  • Long‑context reasoning
  • Large‑scale text generation

Limits

  • No vision support
  • Large VRAM requirement for higher‑bit quantizations
  • License restricts commercial use

Best for

  • Chatbots needing long context
  • Research on instruction tuning
  • Generating detailed technical documents

Who it’s for. Developers and researchers who need a large‑parameter instruction‑tuned model with extensive context. The model is suitable for applications that can accommodate the VRAM demands of higher‑bit quantizations.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S18.2 GB 21.4 GB
Too big Get file →
IQ1_M19.2 GB 22.4 GB
Too big Get file →
IQ2_XXS20.9 GB 24.1 GB
Too big Get file →
IQ2_XS22.5 GB 25.6 GB
Too big Get file →
Q2_K_S24.7 GB 27.9 GB
Too big Get file →
IQ2_S25.8 GB 29.0 GB
Too big Get file →
IQ2_M27.1 GB 30.3 GB
Too big Get file →
IQ3_XXS27.7 GB 30.9 GB
Too big Get file →
IQ3_XS29.4 GB 32.6 GB
Too big Get file →
IQ3_S31.0 GB 34.2 GB
Too big Get file →
Q3_K_S31.0 GB 34.2 GB
Too big Get file →
IQ3_M32.1 GB 35.3 GB
Too big Get file →
Q3_K_M34.4 GB 37.6 GB
Too big Get file →
IQ4_XS37.9 GB 41.1 GB
Too big Get file →
Q4_039.7 GB 42.9 GB
Too big Get file →
IQ4_NL40.1 GB 43.2 GB
Too big Get file →
Q4_K_S40.6 GB 43.8 GB
Too big Get file →
Q4_K_M42.8 GB 46.0 GB
Too big Get file →
Q4_144.1 GB 47.3 GB
Too big Get file →
BF1645.3 GB 48.5 GB
Too big Get file →
F1645.3 GB 48.5 GB
Too big Get file →
Q5_048.5 GB 51.7 GB
Too big Get file →
Q5_K_S48.8 GB 52.0 GB
Too big Get file →
Q5_153.6 GB (2 parts) 56.8 GB
Too big Get file →
Q6_K_M57.9 GB (2 parts) 61.1 GB
Too big Get file →
Q6_K_L58.4 GB (2 parts) 61.6 GB
Too big Get file →
BF1660.4 GB (2 parts) 63.6 GB
Too big Get file →
F1660.4 GB (2 parts) 63.6 GB
Too big Get file →
Q8_075.0 GB (2 parts) 78.2 GB
Too big Get file →
BF1676.9 GB (2 parts) 80.1 GB
Too big Get file →
F1676.9 GB (2 parts) 80.1 GB
Too big Get file →
BF16141.1 GB (4 parts) 144.3 GB
Too big Get file →
F16141.1 GB (4 parts) 144.3 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (18.2 GB file, 21.4 GB memory at 8 k context). For 12 GB VRAM, IQ1_M (19.22 GB file, 22.4 GB memory). For 16 GB VRAM, Q4_K_M (42.79 GB file, 46 GB memory) or Q3_K_M (34.4 GB file, 37.6 GB memory).

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 IQ1_M (19.2 GB file, about 22.4 GB in use).
  • 32 GB VRAM: the best quant that fits is IQ3_XXS (27.7 GB file, about 30.9 GB in use).
  • 48 GB VRAM: the best quant that fits is Q4_1 (44.1 GB file, about 47.3 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 42.8 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 18.2 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
Parameters70.6B
Layers80
Trained context131,072 tokens
KV cache (fp16)320 KB per token

Source

Quantized byMungert
FamilyLlama
Licensellama3.3
Downloads (HF)551
Last updated2025-09-24

Frequently asked questions about Llama-3.3-70B-Instruct

How much VRAM does Llama-3.3-70B-Instruct need?

About 46.0 GB at Q4_K_M with an 8K context window. That is the 42.8 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 Llama-3.3-70B-Instruct on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 21.4 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 18.2 GB, IQ1_M is 19.2 GB, IQ2_XXS is 20.9 GB, IQ2_XS is 22.5 GB, Q2_K_S is 24.7 GB, IQ2_S is 25.8 GB and 27 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 Mungert/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 21.4 GB memory at 8 k context, which exceeds 8 GB VRAM, so it is not feasible.

What is the smallest quantization available?

IQ1_S is the smallest, with a file size of 18.2 GB and an estimated 21.4 GB memory usage at 8 k context.

Does the model support vision tasks?

No, the model has no vision capabilities.

Other GGUFs of Llama-3.1-70B

Similar models

Browse all →

More from Mungert

All by Mungert →

More Llama models

All Llama →

Other large models

All large models →