Instructions to use Ninnix96/G9v3-39A5B-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use Ninnix96/G9v3-39A5B-gguf 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 Ninnix96/G9v3-39A5B-gguf:Q2_K
# Run inference directly in the terminal:
llama cli -hf Ninnix96/G9v3-39A5B-gguf:Q2_K

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Ninnix96/G9v3-39A5B-gguf:Q2_K
# Run inference directly in the terminal:
llama cli -hf Ninnix96/G9v3-39A5B-gguf:Q2_K

Build from source code

git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf Ninnix96/G9v3-39A5B-gguf:Q2_K
# Run inference directly in the terminal:
./build/bin/llama cli -hf Ninnix96/G9v3-39A5B-gguf:Q2_K

Use Docker

docker model run hf.co/Ninnix96/G9v3-39A5B-gguf:Q2_K

Llama new
LM Studio
Jan

vLLM

How to use Ninnix96/G9v3-39A5B-gguf with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Ninnix96/G9v3-39A5B-gguf"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Ninnix96/G9v3-39A5B-gguf",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Ninnix96/G9v3-39A5B-gguf:Q2_K

Ollama
How to use Ninnix96/G9v3-39A5B-gguf with Ollama:
```
ollama run hf.co/Ninnix96/G9v3-39A5B-gguf:Q2_K
```
Unsloth Desktop

How to use Ninnix96/G9v3-39A5B-gguf with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Ninnix96/G9v3-39A5B-gguf:Q2_K

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "Ninnix96/G9v3-39A5B-gguf:Q2_K"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use Ninnix96/G9v3-39A5B-gguf with Docker Model Runner:
```
docker model run hf.co/Ninnix96/G9v3-39A5B-gguf:Q2_K
```

Lemonade

How to use Ninnix96/G9v3-39A5B-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Ninnix96/G9v3-39A5B-gguf:Q2_K

Run and chat with the model

lemonade run user.G9v3-39A5B-gguf-Q2_K

List all available models

lemonade list

Hermes Agent

How to use Ninnix96/G9v3-39A5B-gguf with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Ninnix96/G9v3-39A5B-gguf:Q2_K

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 Ninnix96/G9v3-39A5B-gguf:Q2_K

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use Ninnix96/G9v3-39A5B-gguf with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf Ninnix96/G9v3-39A5B-gguf:Q2_K

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "Ninnix96/G9v3-39A5B-gguf:Q2_K" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

G9v3-39A5B — Q2 GGUF

Fast GGUF for mainstream llama.cpp. Requires a build with G9v3 architecture support (e.g. linuxid10t/llama.cpp@feature/g9v3-support) until upstream llama.cpp merges it — standard builds will not load this model.

Recipe (antirez style)

Tensor group	Format
Routed expert gate / up	IQ2_XXS (imatrix)
Routed expert down	Q2_K
Shared experts, attention, output	Q8_0
Token embeddings	F16
Routers, norms, biases	F32

Same strategy as Salvatore Sanfilippo's DeepSeek V4 quants: imatrix-calibrated IQ2_XXS gate/up with K-quant down routed experts, Q8_0 protection everywhere else. Imatrix: linuxid10t.

Files

File	Size
G9v3-39A5B-Q2-IQ2_XXS-Q2_K.gguf	11.21 GiB

Validation

gguf-dump audit: all routed gate/up are IQ2_XXS, all down are Q2_K (native 256-wide rows, no padding needed), zero misquantized experts; attn/shexp/output Q8_0, token F16, norms/router F32. Spot ppl + KL divergence vs the Q8_0 reference on short prompts (arithmetic, Italian fact, C completion, counting, colors), CPU-only: Q8 PPL 5.21, Q2 PPL 6.77, KL 0.21, same top token 83.9%.

Usage

# needs a G9v3-enabled llama.cpp build, e.g.
# git clone -b feature/g9v3-support https://github.com/linuxid10t/llama.cpp
llama-cli -m G9v3-39A5B-Q2-IQ2_XXS-Q2_K.gguf -t 16 -c 32768

# server
llama-server -m G9v3-39A5B-Q2-IQ2_XXS-Q2_K.gguf -t 16 -c 32768 --host 0.0.0.0 --port 8080

Model

G9v3-39A5B is a 320-expert MoE preview model by AI9Stars. Context: long.

Original model: ai9stars/G9v3-39A5B. Original BF16 GGUF: linuxid10t/G9v3-39A5B-GGUF. License: Apache-2.0.

Downloads last month: 186

GGUF

Model size

39B params

Architecture

g9v3

Hardware compatibility

2-bit

Model tree for Ninnix96/G9v3-39A5B-gguf

Base model

ai9stars/G9v3-39A5B

Quantized

(7)

this model

Collection including Ninnix96/G9v3-39A5B-gguf

Antirez's MoE recipe

Collection

Antirez recipe for MoE quantization: Q2 uses IQ2_XXS gate/up + Q2_K down; Q4 uses Q4_K gate/up + MXFP4 down. Everything else almost unchanged. • 10 items • Updated 9 days ago