Instructions to use efficiencyx/Jun-Lora-v2-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use efficiencyx/Jun-Lora-v2-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="efficiencyx/Jun-Lora-v2-GGUF")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("efficiencyx/Jun-Lora-v2-GGUF")
model = AutoModelForMultimodalLM.from_pretrained("efficiencyx/Jun-Lora-v2-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use efficiencyx/Jun-Lora-v2-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 efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf efficiencyx/Jun-Lora-v2-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 efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf efficiencyx/Jun-Lora-v2-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 efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M

Use Docker

docker model run hf.co/efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use efficiencyx/Jun-Lora-v2-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "efficiencyx/Jun-Lora-v2-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": "efficiencyx/Jun-Lora-v2-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M

SGLang

How to use efficiencyx/Jun-Lora-v2-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "efficiencyx/Jun-Lora-v2-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "efficiencyx/Jun-Lora-v2-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "efficiencyx/Jun-Lora-v2-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "efficiencyx/Jun-Lora-v2-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use efficiencyx/Jun-Lora-v2-GGUF with Ollama:
```
ollama run hf.co/efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M
```
Unsloth Desktop

How to use efficiencyx/Jun-Lora-v2-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M

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": "efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use efficiencyx/Jun-Lora-v2-GGUF with Docker Model Runner:
```
docker model run hf.co/efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M
```

Lemonade

How to use efficiencyx/Jun-Lora-v2-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Jun-Lora-v2-GGUF-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use efficiencyx/Jun-Lora-v2-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 efficiencyx/Jun-Lora-v2-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 efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use efficiencyx/Jun-Lora-v2-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M

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 "efficiencyx/Jun-Lora-v2-GGUF:Q4_K_M" \
  --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"

efficiencyx commited on Jun 26

Commit

20147bd

verified ·

1 Parent(s): 25b46cb

Update README.md

Browse files

Files changed (1) hide show

README.md +3 -8

README.md CHANGED Viewed

@@ -71,7 +71,7 @@ The dataset was constructed to preserve the character's tone, vocabulary, emotio
 | Parameter | Value |
 |:----------|:------|
 | Base model | `google/gemma-4-12b-it` |
-| Method | QLoRA (4-bit NF4) |
 | LoRA rank | 64 |
 | LoRA alpha | 128 |
 | Learning rate | 2e-5 |
@@ -83,11 +83,6 @@ The dataset was constructed to preserve the character's tone, vocabulary, emotio
 | Checkpoint interval | Every 30 steps |
 | Optimizer | AdamW (8-bit) |
-### Why QLoRA instead of LoRA
-This fine-tune uses QLoRA (4-bit NF4 quantization during training) rather than standard full-precision LoRA. The rationale is analogous to Quantization-Aware Training (QAT): by training the adapter on a model that is already quantized to NF4, the learned weights inherently compensate for the precision loss introduced by quantization. This means the adapter is optimized for the same numerical conditions it will encounter at inference time when running on consumer-grade hardware with quantized GGUF models. A LoRA adapter trained at full precision may (and will) underperform when later applied to or merged into a quantized model, because the weight distributions it learned against no longer match the runtime precision.
 ### Infrastructure
 | Component | Detail |
@@ -119,7 +114,7 @@ Multiple adapter checkpoints are provided (steps 90, 120, 138) to allow users to
 ## Acknowledgments
 - **Google** for the [Gemma 4](https://ai.google.dev/gemma) model family
 - **Google Colaboratory** for allowing easy and cheap access to powerful GPU
-- **Unsloth** for the efficient fine-tuning framework
-- **IncontinentCell Studios** for *My Dystopian Robot Girlfriend*, the source material for Jun's character

 | Parameter | Value |
 |:----------|:------|
 | Base model | `google/gemma-4-12b-it` |
+| Method | LoRA |
 | LoRA rank | 64 |
 | LoRA alpha | 128 |
 | Learning rate | 2e-5 |
 | Checkpoint interval | Every 30 steps |
 | Optimizer | AdamW (8-bit) |
 ### Infrastructure
 | Component | Detail |
 ## Acknowledgments
+- **Incontinent Cell** for [*My Dystopian Robot Girlfriend*](https://incontinentcell.itch.io/), Jun's character
 - **Google** for the [Gemma 4](https://ai.google.dev/gemma) model family
 - **Google Colaboratory** for allowing easy and cheap access to powerful GPU
+- **Unsloth** for the efficient fine-tuning framework