Instructions to use QKing-Official/EndAI-Small with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QKing-Official/EndAI-Small with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="QKing-Official/EndAI-Small",
	filename="endai-small-q4.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use QKing-Official/EndAI-Small with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf QKing-Official/EndAI-Small
# Run inference directly in the terminal:
llama-cli -hf QKing-Official/EndAI-Small

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf QKing-Official/EndAI-Small
# Run inference directly in the terminal:
llama-cli -hf QKing-Official/EndAI-Small

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 QKing-Official/EndAI-Small
# Run inference directly in the terminal:
./llama-cli -hf QKing-Official/EndAI-Small

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 QKing-Official/EndAI-Small
# Run inference directly in the terminal:
./build/bin/llama-cli -hf QKing-Official/EndAI-Small

Use Docker

docker model run hf.co/QKing-Official/EndAI-Small

LM Studio
Jan
Ollama
How to use QKing-Official/EndAI-Small with Ollama:
```
ollama run hf.co/QKing-Official/EndAI-Small
```

Unsloth Studio new

How to use QKing-Official/EndAI-Small 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 QKing-Official/EndAI-Small 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 QKing-Official/EndAI-Small to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for QKing-Official/EndAI-Small to start chatting

Docker Model Runner
How to use QKing-Official/EndAI-Small with Docker Model Runner:
```
docker model run hf.co/QKing-Official/EndAI-Small
```

Lemonade

How to use QKing-Official/EndAI-Small with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull QKing-Official/EndAI-Small

Run and chat with the model

lemonade run user.EndAI-Small-{{QUANT_TAG}}

List all available models

lemonade list

QKing-Official commited on 23 days ago

Commit

91d4e30

verified ·

1 Parent(s): 3a92501

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +23 -3

README.md CHANGED Viewed

@@ -1,3 +1,23 @@
----
-license: mit
----

+## EndAI‑small.py (TinyLlama)
+### Purpose
+To quickly finetune a model and easily run and use it with the EndAI server!
+This model is specifically designed by me!
+### Usage
+```bash
+python EndAI-small.py
+```
+### What it does
+1. Downloads `TinyLlama/TinyLlama-1.1B-Chat-v1.0`
+2. Loads 3% of `HuggingFaceH4/ultrachat_200k` (train_sft)
+3. Applies a system prompt and formats.
+4. Tokenises with max length 512
+5. Trains a LoRA adapter (`r=8`, targets `q_proj`, `v_proj`) for 2 epochs
+6. Saves the merged model to `./endai-small-final/`
+Manually convert to gguf using the llama.cpp build from the non-small version of this.