Smol_Llama 4x220M MoE needs about 1.0 GB of memory at Q4_K_M (a 0.4 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 24 quants, from 0.1 GB (IQ1_S) to 0.5 GB (Q6_K).
Smol_Llama 4x220M MoE – a lightweight Mixture of Experts Llama model
Smol_Llama 4x220M MoE is a 0.595 billion‑parameter Llama model with a 2048‑token context window. It is a mixture of experts built from four 220 M sub‑models and supports coding and tools abilities. The model is released under the Apache‑2.0 license.
Good at
- Coding tasks
- Tool integration
- Fast inference on small GPUs
Limits
- No vision support
- Requires 8 k context for memory estimates
- Only GGUF quantizations available
Best for
- Rapid prototyping
- Edge device inference
- Educational demos
Who it’s for. Developers and researchers needing a small, fast Llama variant with expert routing. The model is suitable for local deployment on machines with limited VRAM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 0.1 GB | 0.7 GB | Fits | Get file → | |
| IQ1_M | 0.2 GB | 0.7 GB | Fits | Get file → | |
| IQ2_XXS | 0.2 GB | 0.7 GB | Fits | Get file → | |
| IQ2_XS | 0.2 GB | 0.7 GB | Fits | Get file → | |
| IQ2_S | 0.2 GB | 0.7 GB | Fits | Get file → | |
| IQ2_M | 0.2 GB | 0.7 GB | Fits | Get file → | |
| Q2_K_S | 0.2 GB | 0.7 GB | Fits | Get file → | |
| Q2_K | 0.2 GB | 0.8 GB | Fits | Get file → | |
| IQ3_XXS | 0.2 GB | 0.8 GB | Fits | Get file → | |
| IQ3_XS | 0.3 GB | 0.8 GB | Fits | Get file → | |
| Q3_K_S | 0.3 GB | 0.8 GB | Fits | Get file → | |
| IQ3_S | 0.3 GB | 0.8 GB | Fits | Get file → | |
| IQ3_M | 0.3 GB | 0.8 GB | Fits | Get file → | |
| Q3_K_M | 0.3 GB | 0.8 GB | Fits | Get file → | |
| Q3_K_L | 0.3 GB | 0.8 GB | Fits | Get file → | |
| IQ4_XS | 0.3 GB | 0.8 GB | Fits | Get file → | |
| IQ4_NL | 0.3 GB | 0.9 GB | Fits | Get file → | |
| Q4_0 | 0.3 GB | 0.9 GB | Fits | Get file → | |
| Q4_K_S | 0.3 GB | 0.9 GB | Fits | Get file → | |
| Q4_K_M | 0.4 GB | 0.9 GB | Fits | Get file → | |
| Q4_1 | 0.4 GB | 0.9 GB | Fits | Get file → | |
| Q5_K_S | 0.4 GB | 0.9 GB | Fits | Get file → | |
| Q5_K_M | 0.4 GB | 0.9 GB | Fits | Get file → | |
| Q6_K | 0.5 GB | 1.0 GB | Fits | Get file → |
For 8 GB VRAM, use IQ1_S or IQ1_M (0.14 GB or 0.15 GB). For 12 GB VRAM, Q4_K_M (0.37 GB) or Q5_K_M (0.43 GB) provide a good balance of size and performance. Larger models like Q6_K (0.49 GB) need more than 12 GB VRAM.
What GPU do you need for Smol_Llama 4x220M MoE?
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: the best quant that fits is Q6_K (0.5 GB file, about 1.1 GB in use).
Which Smol_Llama 4x220M MoE quant should you choose?
Q4_K_M is the usual starting point for Smol_Llama 4x220M MoE: it balances file size and quality well. Here it is a 0.4 GB file.
If memory is tight, the smallest file is IQ1_S at 0.1 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 Smol_Llama 4x220M MoE
How much VRAM does Smol_Llama 4x220M MoE need?
About 1.0 GB at Q4_K_M with an 8K context window. That is the 0.4 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 Smol_Llama 4x220M MoE on an 8 GB GPU?
Yes, with Q6_K: the 0.5 GB file needs about 1.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Smol_Llama 4x220M MoE GGUF?
It depends on the quant: IQ1_S is 0.1 GB, IQ1_M is 0.2 GB, IQ2_XXS is 0.2 GB, IQ2_XS is 0.2 GB, IQ2_S is 0.2 GB, IQ2_M is 0.2 GB and 18 more.
What context length does Smol_Llama 4x220M MoE support?
The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.0 GB at 2K.
Where can I download Smol_Llama 4x220M MoE GGUF?
The files are on Hugging Face in the mradermacher/smol_llama-4x220M-MoE-i1-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 Smol_Llama 4x220M MoE use?
Hugging Face lists the license as apache-2.0. 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?
Yes, the IQ1_S and IQ1_M quantizations fit within 8 GB VRAM, using about 0.7 GB of memory at 8 k context.
Does this model support vision tasks?
No, the model has no vision capabilities.
What license applies to this model?
The model is released under the Apache‑2.0 license.