Instructions to use Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF",
	filename="TomGrc_FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S
# Run inference directly in the terminal:
llama-cli -hf Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S
# Run inference directly in the terminal:
llama-cli -hf Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S
# Run inference directly in the terminal:
./llama-cli -hf Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S

Use Docker

docker model run hf.co/Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S

LM Studio
Jan
Ollama
How to use Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF with Ollama:
```
ollama run hf.co/Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S
```

Unsloth Studio new

How to use Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF to start chatting

Docker Model Runner
How to use Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF with Docker Model Runner:
```
docker model run hf.co/Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S
```

Lemonade

How to use Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF:Q4_K_S

Run and chat with the model

lemonade run user.TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF-Q4_K_S

List all available models

lemonade list

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

GGUF Quants with iMatrix for : https://huggingface.co/TomGrc/FusionNet_7Bx2_MoE_v0.1

The second version of a hard-benching model, but without excessive overfit : the perplexity is a bit high, but still drops at longer context before stabilizing, which means that the model is really usable. As far as I know, that's the smartest compromise among the 7bx2 Mistral amateur MOE on the 1st february 2024.

Llama CPP Benchs :

FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,Hellaswag,89,400,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,Hellaswag_Bin,85,400,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,Arc-Challenge,61.20401338,,299,2024-02-03 05:40:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,Arc-Easy,76.31578947,,570,2024-02-03 05:40:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,MMLU,44.40433213,,277,2024-02-03 05:40:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,Thruthful-QA,51.77478580,,817,2024-02-03 05:40:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,Winogrande,83.5833,,1267,2024-02-03 05:40:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,wikitext,6.7506,512,512,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,wikitext,5.4589,4096,4096,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,wikitext,5.3443,6144,6144,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,wikitext,5.5125,7168,7168,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,wikitext,5.2504,8192,8192,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,wikitext,8.9976,10240,10240,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,wikitext,43.9996,12288,12288,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,
FusionNet_7Bx2_MoE_v0.1-b1924-Q8_0.gguf,-,wikitext,780.5238,16384,16384,2024-02-03 00:00:00,,MOE_14b,Mistral_v0.2,8192,,,GGUF,TomGrc,Nexesenex,

Downloads last month: 146

GGUF

Model size

13B params

Architecture

llama

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

8-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Collection including Nexesenex/TomGrc_FusionNet_7Bx2_MoE_v0.1-iMat.GGUF

My favorite models (benchs+usage/feel+innovation)

Collection

How I select models ? First, I bench them, to trim the overfit & dumbified models. Then, I test them. The smartest to my taste end-up here. • 10 items • Updated Mar 19, 2025 • 8