Instructions to use uaytug/fumea-f-dense-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use uaytug/fumea-f-dense-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="uaytug/fumea-f-dense-gguf",
	filename="fumea-f-dense-BF16.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use uaytug/fumea-f-dense-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf uaytug/fumea-f-dense-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf uaytug/fumea-f-dense-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf uaytug/fumea-f-dense-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf uaytug/fumea-f-dense-gguf:Q4_K_M

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 uaytug/fumea-f-dense-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf uaytug/fumea-f-dense-gguf:Q4_K_M

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 uaytug/fumea-f-dense-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf uaytug/fumea-f-dense-gguf:Q4_K_M

Use Docker

docker model run hf.co/uaytug/fumea-f-dense-gguf:Q4_K_M

LM Studio
Jan
Ollama
How to use uaytug/fumea-f-dense-gguf with Ollama:
```
ollama run hf.co/uaytug/fumea-f-dense-gguf:Q4_K_M
```

Unsloth Studio

How to use uaytug/fumea-f-dense-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 uaytug/fumea-f-dense-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 uaytug/fumea-f-dense-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for uaytug/fumea-f-dense-gguf to start chatting

How to use uaytug/fumea-f-dense-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf uaytug/fumea-f-dense-gguf:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "uaytug/fumea-f-dense-gguf:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use uaytug/fumea-f-dense-gguf with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf uaytug/fumea-f-dense-gguf:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default uaytug/fumea-f-dense-gguf:Q4_K_M

Run Hermes

hermes

Docker Model Runner
How to use uaytug/fumea-f-dense-gguf with Docker Model Runner:
```
docker model run hf.co/uaytug/fumea-f-dense-gguf:Q4_K_M
```

Lemonade

How to use uaytug/fumea-f-dense-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull uaytug/fumea-f-dense-gguf:Q4_K_M

Run and chat with the model

lemonade run user.fumea-f-dense-gguf-Q4_K_M

List all available models

lemonade list

FUMEA-F Dense GGUF — Quantized Versions

GGUF quantizations of uaytug/fumea-f-dense — a DARE+TIES merge of 4 specialized Qwen3-8B financial/marketing models.

Architecture

Detail	Value
Base	Qwen3-8B
Merge Method	DARE+TIES (density 0.75)
Context	128K (YaRN RoPE)
Tool Use	Supported

Quantizations

File	Quant	Size	Use Case
fumea-f-dense-F32.gguf	F32	30.52 GB	Research, maximum quality
fumea-f-dense-F16.gguf	F16	15.26 GB	GPU inference, full precision
fumea-f-dense-BF16.gguf	BF16	15.26 GB	GPU inference, bfloat16
fumea-f-dense-Q8_0.gguf	Q8_0	8.11 GB	Near-lossless, high-end GPU/CPU
fumea-f-dense-Q6_K.gguf	Q6_K	6.26 GB	Excellent quality, high-end hardware
fumea-f-dense-Q5_K_M.gguf	Q5_K_M	5.45 GB	Very good quality, balanced
fumea-f-dense-Q5_K_S.gguf	Q5_K_S	5.33 GB	Very good quality, smaller
fumea-f-dense-Q5_0.gguf	Q5_0	5.33 GB	Good quality, moderate hardware
fumea-f-dense-Q4_K_M.gguf	Q4_K_M	4.68 GB	⭐ Recommended — best quality/size
fumea-f-dense-Q4_K_S.gguf	Q4_K_S	4.47 GB	Good quality, smaller footprint
fumea-f-dense-Q4_0.gguf	Q4_0	4.45 GB	Decent quality, broad compatibility
fumea-f-dense-Q3_K_L.gguf	Q3_K_L	4.13 GB	Acceptable quality, lower memory
fumea-f-dense-Q3_K_M.gguf	Q3_K_M	3.84 GB	Lower quality, memory constrained
fumea-f-dense-Q3_K_S.gguf	Q3_K_S	3.51 GB	Lower quality, very constrained
fumea-f-dense-Q2_K.gguf	Q2_K	3.06 GB	Minimal quality, extreme compression
fumea-f-dense-IQ4_XS.gguf	IQ4_XS	4.28 GB	imatrix 4-bit extra small
fumea-f-dense-IQ4_NL.gguf	IQ4_NL	4.49 GB	Non-linear 4-bit
fumea-f-dense-IQ3_M.gguf	IQ3_M	3.63 GB	imatrix 3-bit medium
fumea-f-dense-IQ3_S.gguf	IQ3_S	3.53 GB	imatrix 3-bit small
fumea-f-dense-IQ3_XXS.gguf	IQ3_XXS	0.65 GB	imatrix 3-bit XXS
fumea-f-dense-IQ2_M.gguf	IQ2_M	0.65 GB	Extreme 2-bit medium

Recommended: Q4_K_M for the best quality/size balance.

Ollama

wget https://huggingface.co/uaytug/fumea-f-dense-gguf/resolve/main/fumea-f-dense-Q4_K_M.gguf

cat > Modelfile << 'EOF'
FROM ./fumea-f-dense-Q4_K_M.gguf
PARAMETER temperature 0.6
PARAMETER top_p 0.9
PARAMETER repeat_penalty 1.1
PARAMETER num_ctx 32768
SYSTEM You are FUMEA-F, a financial analysis and marketing intelligence assistant.
EOF

ollama create fumea-f-dense -f Modelfile
ollama run fumea-f-dense

LM Studio

Download any GGUF file from this repo
LM Studio → My Models → Import
Start chatting

License

Apache 2.0

Downloads last month: 133

GGUF

Model size

8B params

Architecture

qwen3

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

16-bit

32-bit

Inference Providers NEW

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

Model tree for uaytug/fumea-f-dense-gguf

Base model

uaytug/fumea-f-dense

Quantized

(1)

this model

Collection including uaytug/fumea-f-dense-gguf

FUMEA-F

Collection

FUMEA-F(Frontier Unified Multi-Expert Agent - Financial) is a Marketing and Finance Focused model • 3 items • Updated Apr 10

uaytug
/

fumea-f-dense-gguf

FUMEA-F Dense GGUF — Quantized Versions

Architecture

Quantizations

Ollama

LM Studio

See Also

License

Model tree for uaytug/fumea-f-dense-gguf

Collection including uaytug/fumea-f-dense-gguf

FUMEA-F