FusionNet_34Bx2_MoE needs about 39.2 GB of memory at Q4_K_M (a 36.7 GB file) with an 8K context window, so it fits on a 48 GB GPU. This repository has 21 quants, from 12.8 GB (IQ1_S) to 49.9 GB (Q6_K). Longer context needs more: at 32K tokens the KV cache alone adds about 8.1 GB.
FusionNet_34Bx2_MoE – 60.8B parameter Llama model with Mixture of Experts
FusionNet_34Bx2_MoE is a 60.8B parameter Llama architecture fine‑tuned on English using MoE. It supports a 32768 token context window and is released under the MIT license. The model is available in multiple GGUF quantizations.
Good at
- Large context generation
- Mixture of Experts efficiency
- English fine‑tuned responses
Limits
- No vision support
- Requires 8k context for memory estimates
- MIT license only
Best for
- Long‑form content creation
- Complex Q&A sessions
- Research prototyping
Who it’s for. Developers and researchers needing a large Llama model with MoE for English language tasks. The model is suitable for experimentation and fine‑tuning in local environments.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 12.8 GB | 15.3 GB | Too big | Get file → | |
| IQ1_M | 14.1 GB | 16.6 GB | Too big | Get file → | |
| IQ2_XXS | 16.2 GB | 18.7 GB | Too big | Get file → | |
| IQ2_XS | 18.0 GB | 20.5 GB | Too big | Get file → | |
| IQ2_S | 18.7 GB | 21.2 GB | Too big | Get file → | |
| IQ2_M | 20.4 GB | 22.9 GB | Too big | Get file → | |
| Q2_K | 22.4 GB | 24.9 GB | Too big | Get file → | |
| IQ3_XXS | 23.5 GB | 26.0 GB | Too big | Get file → | |
| IQ3_XS | 25.0 GB | 27.5 GB | Too big | Get file → | |
| Q3_K_S | 26.3 GB | 28.8 GB | Too big | Get file → | |
| IQ3_S | 26.4 GB | 28.9 GB | Too big | Get file → | |
| IQ3_M | 27.1 GB | 29.6 GB | Too big | Get file → | |
| Q3_K_M | 29.2 GB | 31.8 GB | Too big | Get file → | |
| Q3_K_L | 31.8 GB | 34.3 GB | Too big | Get file → | |
| IQ4_XS | 32.5 GB | 35.0 GB | Too big | Get file → | |
| Q4_0 | 34.5 GB | 37.0 GB | Too big | Get file → | |
| Q4_K_S | 34.6 GB | 37.1 GB | Too big | Get file → | |
| Q4_K_M | 36.7 GB | 39.2 GB | Too big | Get file → | |
| Q5_K_S | 41.9 GB | 44.4 GB | Too big | Get file → | |
| Q5_K_M | 43.1 GB | 45.6 GB | Too big | Get file → | |
| Q6_K | 49.9 GB | 52.4 GB | Too big | Get file → |
For 8GB VRAM, use IQ1_S (12.8 GB file, 15.3 GB memory at 8k context). For 12GB VRAM, IQ1_M (14.07 GB file, 16.6 GB memory). For 16GB VRAM, IQ2_XS (17.99 GB file, 20.5 GB memory).
What GPU do you need for FusionNet_34Bx2_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: 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 IQ1_S (12.8 GB file, about 15.3 GB in use).
- 24 GB VRAM: the best quant that fits is IQ2_M (20.4 GB file, about 22.9 GB in use).
- 32 GB VRAM: the best quant that fits is Q3_K_M (29.2 GB file, about 31.8 GB in use).
- 48 GB VRAM: the best quant that fits is Q5_K_M (43.1 GB file, about 45.6 GB in use).
Which FusionNet_34Bx2_MoE quant should you choose?
Q4_K_M is the usual starting point for FusionNet_34Bx2_MoE: it balances file size and quality well. Here it is a 36.7 GB file.
If memory is tight, the smallest file is IQ1_S at 12.8 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_34Bx2_MoE
How much VRAM does FusionNet_34Bx2_MoE need?
About 39.2 GB at Q4_K_M with an 8K context window. That is the 36.7 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_34Bx2_MoE on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 15.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of FusionNet_34Bx2_MoE GGUF?
It depends on the quant: IQ1_S is 12.8 GB, IQ1_M is 14.1 GB, IQ2_XXS is 16.2 GB, IQ2_XS is 18.0 GB, IQ2_S is 18.7 GB, IQ2_M is 20.4 GB and 15 more.
What context length does FusionNet_34Bx2_MoE support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 2.0 GB at 8K and 8.1 GB at 32K.
Where can I download FusionNet_34Bx2_MoE GGUF?
The files are on Hugging Face in the mradermacher/FusionNet_34Bx2_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 FusionNet_34Bx2_MoE 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_34Bx2_MoE on 8GB VRAM?
The IQ1_S quant requires 15.3 GB memory at 8k context, so 8GB VRAM is insufficient.
Does the model support image inputs?
No, FusionNet_34Bx2_MoE has no vision capabilities.
What license applies to this model?
The model is released under the MIT license.