Instructions to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with Transformers:

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

pipe = pipeline("text-generation", model="0xA50C1A1/Llama-3.3-8B-Nymphaea-RP")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("0xA50C1A1/Llama-3.3-8B-Nymphaea-RP")
model = AutoModelForCausalLM.from_pretrained("0xA50C1A1/Llama-3.3-8B-Nymphaea-RP")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

llama-cpp-python

How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="0xA50C1A1/Llama-3.3-8B-Nymphaea-RP",
	filename="Llama-3.3-8B-Nymphaea-RP.Q5_Mix.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP
# Run inference directly in the terminal:
llama-cli -hf 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP
# Run inference directly in the terminal:
llama-cli -hf 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

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 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP
# Run inference directly in the terminal:
./llama-cli -hf 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

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 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP
# Run inference directly in the terminal:
./build/bin/llama-cli -hf 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

Use Docker

docker model run hf.co/0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

LM Studio
Jan

vLLM

How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "0xA50C1A1/Llama-3.3-8B-Nymphaea-RP"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "0xA50C1A1/Llama-3.3-8B-Nymphaea-RP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

SGLang

How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP 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 "0xA50C1A1/Llama-3.3-8B-Nymphaea-RP" \
    --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": "0xA50C1A1/Llama-3.3-8B-Nymphaea-RP",
		"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 "0xA50C1A1/Llama-3.3-8B-Nymphaea-RP" \
        --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": "0xA50C1A1/Llama-3.3-8B-Nymphaea-RP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with Ollama:
```
ollama run hf.co/0xA50C1A1/Llama-3.3-8B-Nymphaea-RP
```

Unsloth Studio

How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP 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 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP 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 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP to start chatting

How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

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": "0xA50C1A1/Llama-3.3-8B-Nymphaea-RP"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

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 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

Run Hermes

hermes

Docker Model Runner
How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with Docker Model Runner:
```
docker model run hf.co/0xA50C1A1/Llama-3.3-8B-Nymphaea-RP
```

Lemonade

How to use 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull 0xA50C1A1/Llama-3.3-8B-Nymphaea-RP

Run and chat with the model

lemonade run user.Llama-3.3-8B-Nymphaea-RP-{{QUANT_TAG}}

List all available models

lemonade list

Llama-3.3-8B-Nymphaea-RP

A fine-tune of Llama 3.3 8B Instruct for roleplay and creative writing.

I've trained this mostly for merging with Llama 3.1/3.3 8B fine-tunes.

The SillyTavern preset is available here. For custom presets, please use the Llama 3 instruct template.

GGUF

Here is my custom mixed-quant GGUF, optimized for 8/12GB VRAM.

GGUF recipe

llama-quantize \
  --imatrix imatrix.gguf \
  --token-embedding-type q8_0 \
  --output-tensor-type q8_0 \
  --tensor-type ".*attn_q.weight=q8_0" \
  --tensor-type ".*attn_k.weight=q8_0" \
  --tensor-type ".*attn_v.weight=q5_k" \
  --tensor-type ".*attn_output.weight=q5_k" \
  --tensor-type ".*ffn_up.weight=iq4_nl" \
  --tensor-type ".*ffn_gate.weight=iq4_nl" \
  --tensor-type ".*ffn_down.weight=q5_k" \
  Llama-3.3-8B-Nymphaea-RP.F16.gguf \
  Llama-3.3-8B-Nymphaea-RP.Q5_Mix.gguf \
  q5_k

Imatrix file for making your own quants is available here. I used this calibration dataset to create it, expanding it with RP and creative writing data (about 400k tokens).

Training Notes

Trained on the latest iteration of my Darkmere dataset. This version features expanded genre variety, built upon a mix of manually curated synthetics and human-written stories.

The base weights are abliterated via Heretic prior to fine-tuning, so this fine-tune is quite uncensored.

Training Specs

Method:

Training Method: DoRA (Weight-Decomposed LoRA)
Target Modules all-linear
LoRA Rank: 64
LoRA Alpha: 64
LoRA Dropout: 0.05

Hyperparameters:

Batch Size: 2 (Per-device)
Gradient Accumulation: 2
Epochs: 2
Learning Rate: 1e-4
Optimizer: adamw_torch_fused
LR Scheduler: cosine
Noise Level: neftune_noise_alpha=5

Special Thanks

This fine-tune wouldn't be possible without the incredible work of the community:

p-e-w for developing Heretic - an essential tool for censorship removal.
SicariusSicariiStuff for developing SLOP_Detector script.
allura-forge and shb777 for providing access to the Llama 3.3 8B weights.
AMD for their Instinct™ MI300X GPU.