Instructions to use ProjectMosiacAI/Hades-8B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use ProjectMosiacAI/Hades-8B with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ProjectMosiacAI/Hades-8B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ProjectMosiacAI/Hades-8B:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ProjectMosiacAI/Hades-8B:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf ProjectMosiacAI/Hades-8B: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 ProjectMosiacAI/Hades-8B:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf ProjectMosiacAI/Hades-8B: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 ProjectMosiacAI/Hades-8B:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ProjectMosiacAI/Hades-8B:Q4_K_M

Use Docker

docker model run hf.co/ProjectMosiacAI/Hades-8B:Q4_K_M

LM Studio
Jan

vLLM

How to use ProjectMosiacAI/Hades-8B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ProjectMosiacAI/Hades-8B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ProjectMosiacAI/Hades-8B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ProjectMosiacAI/Hades-8B:Q4_K_M

Ollama
How to use ProjectMosiacAI/Hades-8B with Ollama:
```
ollama run hf.co/ProjectMosiacAI/Hades-8B:Q4_K_M
```
Unsloth Desktop
Docker Model Runner
How to use ProjectMosiacAI/Hades-8B with Docker Model Runner:
```
docker model run hf.co/ProjectMosiacAI/Hades-8B:Q4_K_M
```

Lemonade

How to use ProjectMosiacAI/Hades-8B with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ProjectMosiacAI/Hades-8B:Q4_K_M

Run and chat with the model

lemonade run user.Hades-8B-Q4_K_M

List all available models

lemonade list

Atomic Chat

Hades 8B (v2.0) — Small Language Model (SLM)

Hades 8B is an open-source Small Language Model built on Meta's Llama 3.1 8B architecture. Post-trained across 50,000 steps on C-DAC Airawat supercomputing infrastructure using the mlabonne/FineTome-100k dataset, it is optimized for high-reasoning instruction following, structured output, and fast local execution.

Empirical MMLU Benchmark Results

Evaluated using lm-evaluation-harness (lm-eval) in Zero-Shot Chain-of-Thought mode:

Metric / Task	Score	Benchmark Notes
MMLU Overall Macro Average	68.4%	Macro average across 57 standard MMLU subjects
Management	90.9%	Outperforms standard 8B base model baselines
Medical Genetics	81.8%	Specialized biological domain performance
Philosophy	73.5%	Conceptual reasoning proficiency
High School Psychology	70.0%	Behavioral science evaluation
Elementary Mathematics	68.3%	Quantitative baseline

Verified Capabilities

Instruction Following & Alignment: FineTome-100k trained for coherent, multi-turn conversational responses.
Structured Output: Optimized for JSON schema adherence and tool-calling structures.
Edge-Ready Execution: Packaged in 4-bit Q4_K_M GGUF format (4.92 GB), running smoothly on local consumer GPUs/RAM (<6GB VRAM).

Quick Start with Ollama

Run locally via Ollama:

ollama run hf.co/ProjectMosiacAI/Hades-8B

Downloads last month: 53

GGUF

Model size

8B params

Architecture

llama

Hardware compatibility

4-bit

Model tree for ProjectMosiacAI/Hades-8B

Base model

meta-llama/Llama-3.1-8B

Finetuned

meta-llama/Llama-3.1-8B-Instruct

Quantized

(912)

this model

ProjectMosiacAI
/

Hades-8B

Hades 8B (v2.0) — Small Language Model (SLM)

Empirical MMLU Benchmark Results

Verified Capabilities

Quick Start with Ollama

Model tree for ProjectMosiacAI/Hades-8B

Dataset used to train ProjectMosiacAI/Hades-8B