Llama 4 Scout 17B 6E Instruct needs about 30.9 GB of memory at Q4_K_M (a 28.8 GB file) with an 8K context window, so it fits on a 32 GB GPU. This repository has 23 quants, from 10.2 GB (IQ1_S) to 38.9 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.
Llama 4 Scout 17B 6E Instruct – a 47.4 B parameter MoE model with 6‑expert fusion
Llama 4 Scout 17B 6E Instruct is a 47.4 B parameter mixture‑of‑experts model with 6 experts. It supports text and image input and is optimized for instruction following. The model was trained on a 10M token context length and is available in multiple GGUF quantizations.
Good at
- Instruction following
- Image‑aware text generation
- Tool‑use capability
Limits
- Requires 8 k context for memory estimates
- Large VRAM needed for higher‑bit quants
- No explicit license information
Best for
- Local instruction‑driven chatbots
- Image‑guided content creation
- Tool‑integrated automation
Who it’s for. Developers building local LLM services that need multimodal input and instruction following. The model is suitable for users with sufficient GPU memory to run higher‑bit quantizations.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 10.2 GB | 12.3 GB | Too big | Get file → | |
| IQ1_M | 11.2 GB | 13.3 GB | Too big | Get file → | |
| IQ2_XXS | 12.9 GB | 15.0 GB | Too big | Get file → | |
| IQ2_XS | 14.3 GB | 16.4 GB | Too big | Get file → | |
| IQ2_S | 14.6 GB | 16.8 GB | Too big | Get file → | |
| IQ2_M | 16.0 GB | 18.1 GB | Too big | Get file → | |
| Q2_K_S | 16.5 GB | 18.7 GB | Too big | Get file → | |
| Q2_K | 17.7 GB | 19.8 GB | Too big | Get file → | |
| IQ3_XXS | 18.5 GB | 20.6 GB | Too big | Get file → | |
| IQ3_XS | 19.8 GB | 21.9 GB | Too big | Get file → | |
| Q3_K_S | 20.8 GB | 22.9 GB | Too big | Get file → | |
| IQ3_S | 20.8 GB | 22.9 GB | Too big | Get file → | |
| IQ3_M | 21.2 GB | 23.3 GB | Too big | Get file → | |
| Q3_K_M | 23.0 GB | 25.1 GB | Too big | Get file → | |
| Q3_K_L | 24.8 GB | 26.9 GB | Too big | Get file → | |
| IQ4_XS | 25.5 GB | 27.6 GB | Too big | Get file → | |
| Q4_0 | 27.0 GB | 29.2 GB | Too big | Get file → | |
| Q4_K_S | 27.2 GB | 29.3 GB | Too big | Get file → | |
| Q4_K_M | 28.8 GB | 30.9 GB | Too big | Get file → | |
| Q4_1 | 29.8 GB | 31.9 GB | Too big | Get file → | |
| Q5_K_S | 32.7 GB | 34.8 GB | Too big | Get file → | |
| Q5_K_M | 33.7 GB | 35.8 GB | Too big | Get file → | |
| Q6_K | 38.9 GB | 41.0 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (10.21 GB file, 12.3 GB memory at 8k context). For 12 GB VRAM, IQ1_M (11.21 GB file, 13.3 GB memory). For 16 GB VRAM, IQ2_XS (14.27 GB file, 16.4 GB memory).
What GPU do you need for Llama 4 Scout 17B 6E 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: the best quant that fits is IQ2_XXS (12.9 GB file, about 15.0 GB in use).
- 24 GB VRAM: the best quant that fits is IQ3_M (21.2 GB file, about 23.3 GB in use).
- 32 GB VRAM: the best quant that fits is Q4_1 (29.8 GB file, about 31.9 GB in use).
- 48 GB VRAM: the best quant that fits is Q6_K (38.9 GB file, about 41.0 GB in use).
Which Llama 4 Scout 17B 6E Instruct quant should you choose?
Q4_K_M is the usual starting point for Llama 4 Scout 17B 6E Instruct: it balances file size and quality well. Here it is a 28.8 GB file.
If memory is tight, the smallest file is IQ1_S at 10.2 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
Frequently asked questions about Llama 4 Scout 17B 6E Instruct
How much VRAM does Llama 4 Scout 17B 6E Instruct need?
About 30.9 GB at Q4_K_M with an 8K context window. That is the 28.8 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 4 Scout 17B 6E Instruct on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 12.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Llama 4 Scout 17B 6E Instruct GGUF?
It depends on the quant: IQ1_S is 10.2 GB, IQ1_M is 11.2 GB, IQ2_XXS is 12.9 GB, IQ2_XS is 14.3 GB, IQ2_S is 14.6 GB, IQ2_M is 16.0 GB and 17 more.
What context length does Llama 4 Scout 17B 6E Instruct support?
The GGUF header lists a trained context length of 10,485,760 tokens (10240K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.
Where can I download Llama 4 Scout 17B 6E Instruct GGUF?
The files are on Hugging Face in the mradermacher/Llama-4-Scout-17B-6E-Instruct-i1-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
Can I run Llama 4 Scout 17B 6E Instruct on an RTX 4090 with 8 GB VRAM?
Only the lowest‑bit quants such as IQ1_S may fit, but memory usage is still close to the limit. Higher‑bit quants will exceed the VRAM.
Does the model support image input?
Yes, it accepts image input and outputs text and code.
What context length does the model support?
The model was trained with a 10M token context length.