Llama 3.3 70B Instruct DASHQ Q2 needs about 24.8 GB of memory at IQ2_XS (a 21.6 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 4 quants, from 19.4 GB (IQ2_XXS) to 26.6 GB (Q2_K_XL). Longer context needs more: at 32K tokens the KV cache alone adds about 10.7 GB.
Llama 3.3 70B Instruct quantized to 2‑bit with DASH‑Q
Llama 3.3 70B Instruct is a large language model with 70.6 billion parameters. It is quantized to 2‑bit using DASH‑Q, providing several file sizes. The model supports tool usage and has a 131072‑token context window.
Good at
- Tool integration
- Large context handling
- Low‑bit inference
Limits
- No vision support
- Requires large VRAM for higher quants
- License is llama3.3
Best for
- Chatbots needing 131072‑token context
- Tool‑enabled assistants
- Low‑memory inference on GPUs
Who it’s for. Developers and researchers who need a 70B model with 2‑bit quantization for efficient inference. Suitable for projects that can handle the model’s memory demands and require tool‑based interactions.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ2_XXS | 19.4 GB | 22.5 GB | Too big | Get file → | |
| IQ2_XS | 21.6 GB | 24.8 GB | Too big | Get file → | |
| IQ2_M | 24.2 GB | 27.4 GB | Too big | Get file → | |
| Q2_K_XL | 26.6 GB | 29.8 GB | Too big | Get file → |
For 8 GB VRAM, use IQ2_XXS (19.36 GB file, ~22.5 GB memory). For 12 GB VRAM, IQ2_XS (21.63 GB file, ~24.8 GB memory). For 16 GB VRAM, IQ2_M (24.22 GB file, ~27.4 GB memory). Higher quants need more VRAM.
What GPU do you need for Llama 3.3 70B Instruct DASHQ Q2?
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 22.5 GB in use).
- 32 GB VRAM: the best quant that fits is Q2_K_XL (26.6 GB file, about 29.8 GB in use).
Which Llama 3.3 70B Instruct DASHQ Q2 quant should you choose?
IQ2_XS (21.6 GB) is the middle option in this repository. 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.
If memory is tight, the smallest file is IQ2_XXS at 19.4 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 DASHQ Q2
How much VRAM does Llama 3.3 70B Instruct DASHQ Q2 need?
About 24.8 GB at IQ2_XS with an 8K context window. That is the 21.6 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 DASHQ Q2 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 22.5 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Llama 3.3 70B Instruct DASHQ Q2 GGUF?
It depends on the quant: IQ2_XXS is 19.4 GB, IQ2_XS is 21.6 GB, IQ2_M is 24.2 GB, Q2_K_XL is 26.6 GB.
What context length does Llama 3.3 70B Instruct DASHQ Q2 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 DASHQ Q2 GGUF?
The files are on Hugging Face in the jkim96/Llama-3.3-70B-Instruct-DASHQ-Q2-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 DASHQ Q2 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 this model on 8 GB VRAM?
Use the IQ2_XXS quant, which requires about 22.5 GB of VRAM at 8k context. It is the smallest available quant for this model.
Does this model support vision tasks?
No, the model does not include vision capabilities.
What license does this model use?
It inherits the llama3.3 license from the base model.