Instructions to use QuantFactory/Ninja-V2-7B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QuantFactory/Ninja-V2-7B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="QuantFactory/Ninja-V2-7B-GGUF",
	filename="Ninja-V2-7B.Q2_K.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use QuantFactory/Ninja-V2-7B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf QuantFactory/Ninja-V2-7B-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 QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf QuantFactory/Ninja-V2-7B-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 QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf QuantFactory/Ninja-V2-7B-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 QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use QuantFactory/Ninja-V2-7B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QuantFactory/Ninja-V2-7B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QuantFactory/Ninja-V2-7B-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M

Ollama
How to use QuantFactory/Ninja-V2-7B-GGUF with Ollama:
```
ollama run hf.co/QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M
```

Unsloth Studio new

How to use QuantFactory/Ninja-V2-7B-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 QuantFactory/Ninja-V2-7B-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 QuantFactory/Ninja-V2-7B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for QuantFactory/Ninja-V2-7B-GGUF to start chatting

Docker Model Runner
How to use QuantFactory/Ninja-V2-7B-GGUF with Docker Model Runner:
```
docker model run hf.co/QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M
```

Lemonade

How to use QuantFactory/Ninja-V2-7B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull QuantFactory/Ninja-V2-7B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Ninja-V2-7B-GGUF-Q4_K_M

List all available models

lemonade list

Ninja-V2-7B-GGUF

This is quantized version of Local-Novel-LLM-project/Ninja-V2-7B created using llama.cpp

Model Description

このモデルは、ベクトルマージなどを用い作成された高性能ベースモデルです。用途はチャットのみならず、文章創作など幅広いタスクに対応できます。

このモデルは、ローカルLLMに向き合う会企画のLocalAIハッカソンの高性能GPUサーバーの助けを借りて作成されました。

関係者の皆様に感謝申し上げます。

作成方法

モデルレシピは下記の通りです。

Ninja-v2-Base (mergekitにて作成)

models:
  - model: MTSAIR/multi_verse_model
  - model: HuggingFaceH4/zephyr-7b-beta
merge_method: model_stock
base_model: amazingvince/Not-WizardLM-2-7B
dtype: bfloat16

Novels-7B(ninja_mergerにて作成)

target_model: "stabilityai/japanese-stablelm-instruct-gamma-7b"
  -
    left:  "Elizezen/Phos-7B"          # ベースモデルの指定
    right: "stabilityai/japanese-stablelm-instruct-gamma-7b"        # サブモデルの指定
    operation: "sub"                             # 組み合わせの操作。"mix"、"add"などを指定
    velocity: 1.0

  -
    left:  "Elizezen/Antler-7B"          # ベースモデルの指定
    right: "stabilityai/japanese-stablelm-instruct-gamma-7b"        # サブモデルの指定
    operation: "sub"                             # 組み合わせの操作。"mix"、"add"などを指定
    velocity: 1.0

Ninja-v2(ninja_mergerにて作成)

target_model: "Ninja-v2-Base"
models:                                     # 組み合わせの重み。0.0から1.0の範囲で指定
  -
    left:  "NTQAI/chatntq-ja-7b-v1.0"          # ベースモデルの指定
    right: "mistralai/Mistral-7B-v0.1"        # サブモデルの指定0
    operation: "sub"                        # 組み合わせの操作。"mix"、"add"などを指定
    velocity: 1.0
  -
    left:  "Elizezen/Berghof-NSFW-7B"          # ベースモデルの指定
    right: "stabilityai/japanese-stablelm-instruct-gamma-7b"        # サブモデルの指定
    operation: "sub"                             # 組み合わせの操作。"mix"、"add"などを指定
    velocity: 0.5
  -
    left:  "Novels-7B"          # ベースモデルの指定
    right: "stabilityai/japanese-stablelm-instruct-gamma-7b"        # サブモデルの指定
    operation: "sub"                             # 組み合わせの操作。"mix"、"add"などを指定
    velocity: 1.0

プロンプトテンプレート

必須ではありませんが、Vicuna-1.1テンプレートを使用することができます。単純な文章生成においては推奨のテンプレートはありません。

システムプロンプト

BAD:　あなたは○○として振る舞います
GOOD: あなたは○○です
BAD: あなたは○○ができます
GOOD: あなたは○○をします

制限

Apache-2.0ライセンスに従い使用してください。

Downloads last month: 83

GGUF

Model size

7B params

Architecture

llama

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Model tree for QuantFactory/Ninja-V2-7B-GGUF

Base model

Local-Novel-LLM-project/Ninja-V2-7B

Quantized

(3)

this model