FusionNet needs about 16.5 GB of memory at Q4_K_M (a 12.7 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 21 quants, from 4.6 GB (IQ1_S) to 17.4 GB (Q6_K).
FusionNet 21.2B Llama model fine‑tuned with passthrough Fusion
FusionNet is a 21.2B parameter Llama architecture model fine‑tuned on English using the passthrough Fusion method. It supports a 4096 token context window and is released under the MIT license.
Good at
- Fine‑tuned English text generation
- Large context handling
- Versatile quantization options
Limits
- No vision support
- Requires 8k context for memory estimates
- No explicit language support list
Best for
- Chatbot development
- Content drafting
- Research prototyping
Who it’s for. Developers and researchers needing a large, fine‑tuned Llama model with flexible quantization. The model is open source and can be run locally with llama.cpp.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 4.6 GB | 6.7 GB | Fits | Get file → | |
| IQ1_M | 5.0 GB | 7.1 GB | Fits | Get file → | |
| IQ2_XXS | 5.7 GB | 7.8 GB | Tight | Get file → | |
| IQ2_XS | 6.3 GB | 8.4 GB | Too big | Get file → | |
| IQ2_S | 6.6 GB | 8.7 GB | Too big | Get file → | |
| IQ2_M | 7.2 GB | 9.3 GB | Too big | Get file → | |
| Q2_K | 7.9 GB | 10.0 GB | Too big | Get file → | |
| IQ3_XXS | 8.2 GB | 10.3 GB | Too big | Get file → | |
| IQ3_XS | 8.7 GB | 10.8 GB | Too big | Get file → | |
| Q3_K_S | 9.2 GB | 11.3 GB | Too big | Get file → | |
| IQ3_S | 9.2 GB | 11.3 GB | Too big | Get file → | |
| IQ3_M | 9.5 GB | 11.6 GB | Too big | Get file → | |
| Q3_K_M | 10.2 GB | 12.3 GB | Too big | Get file → | |
| Q3_K_L | 11.1 GB | 13.2 GB | Too big | Get file → | |
| IQ4_XS | 11.4 GB | 13.5 GB | Too big | Get file → | |
| Q4_0 | 12.0 GB | 14.1 GB | Too big | Get file → | |
| Q4_K_S | 12.1 GB | 14.2 GB | Too big | Get file → | |
| Q4_K_M | 12.7 GB | 14.9 GB | Too big | Get file → | |
| Q5_K_S | 14.6 GB | 16.7 GB | Too big | Get file → | |
| Q5_K_M | 15.0 GB | 17.1 GB | Too big | Get file → | |
| Q6_K | 17.4 GB | 19.5 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (4.57 GB, 8.3 GB memory) or IQ1_M (5 GB, 8.7 GB memory). For 12 GB VRAM, IQ2_S (6.64 GB, 10.4 GB memory) or Q4_K_M (12.74 GB, 16.5 GB memory) are suitable. Memory estimates are approximate.
What GPU do you need for FusionNet?
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: the best quant that fits is IQ3_XXS (8.2 GB file, about 11.9 GB in use).
- 16 GB VRAM: the best quant that fits is Q4_K_S (12.1 GB file, about 15.8 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K (17.4 GB file, about 21.1 GB in use).
Which FusionNet quant should you choose?
Q4_K_M is the usual starting point for FusionNet: it balances file size and quality well. Here it is a 12.7 GB file.
If memory is tight, the smallest file is IQ1_S at 4.6 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 FusionNet
How much VRAM does FusionNet need?
About 16.5 GB at Q4_K_M with an 8K context window. That is the 12.7 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 FusionNet on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 8.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of FusionNet GGUF?
It depends on the quant: IQ1_S is 4.6 GB, IQ1_M is 5.0 GB, IQ2_XXS is 5.7 GB, IQ2_XS is 6.3 GB, IQ2_S is 6.6 GB, IQ2_M is 7.2 GB and 15 more.
What context length does FusionNet support?
The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 3.2 GB at 8K and 1.6 GB at 4K.
Where can I download FusionNet GGUF?
The files are on Hugging Face in the mradermacher/FusionNet_passthrough-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 FusionNet use?
Hugging Face lists the license as mit. 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 FusionNet on 8 GB VRAM?
Yes, the IQ1_S and IQ1_M quantizations fit within 8 GB VRAM, using about 8.3–8.7 GB memory at 8k context.
Does FusionNet support vision tasks?
No, the model has no vision capabilities.