Instructions to use konpep/Aether-Coder-v2.0-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use konpep/Aether-Coder-v2.0-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="konpep/Aether-Coder-v2.0-GGUF")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("konpep/Aether-Coder-v2.0-GGUF", dtype="auto")

llama-cpp-python

How to use konpep/Aether-Coder-v2.0-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="konpep/Aether-Coder-v2.0-GGUF",
	filename="Aether_coder_v2.0.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 konpep/Aether-Coder-v2.0-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf konpep/Aether-Coder-v2.0-GGUF
# Run inference directly in the terminal:
llama-cli -hf konpep/Aether-Coder-v2.0-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf konpep/Aether-Coder-v2.0-GGUF
# Run inference directly in the terminal:
llama-cli -hf konpep/Aether-Coder-v2.0-GGUF

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 konpep/Aether-Coder-v2.0-GGUF
# Run inference directly in the terminal:
./llama-cli -hf konpep/Aether-Coder-v2.0-GGUF

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 konpep/Aether-Coder-v2.0-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf konpep/Aether-Coder-v2.0-GGUF

Use Docker

docker model run hf.co/konpep/Aether-Coder-v2.0-GGUF

LM Studio
Jan

vLLM

How to use konpep/Aether-Coder-v2.0-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "konpep/Aether-Coder-v2.0-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "konpep/Aether-Coder-v2.0-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/konpep/Aether-Coder-v2.0-GGUF

SGLang

How to use konpep/Aether-Coder-v2.0-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 "konpep/Aether-Coder-v2.0-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "konpep/Aether-Coder-v2.0-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "konpep/Aether-Coder-v2.0-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "konpep/Aether-Coder-v2.0-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use konpep/Aether-Coder-v2.0-GGUF with Ollama:
```
ollama run hf.co/konpep/Aether-Coder-v2.0-GGUF
```

Unsloth Studio new

How to use konpep/Aether-Coder-v2.0-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 konpep/Aether-Coder-v2.0-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 konpep/Aether-Coder-v2.0-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for konpep/Aether-Coder-v2.0-GGUF to start chatting

Docker Model Runner
How to use konpep/Aether-Coder-v2.0-GGUF with Docker Model Runner:
```
docker model run hf.co/konpep/Aether-Coder-v2.0-GGUF
```

Lemonade

How to use konpep/Aether-Coder-v2.0-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull konpep/Aether-Coder-v2.0-GGUF

Run and chat with the model

lemonade run user.Aether-Coder-v2.0-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

🌌 Aether Coder v2.0 (Uncensored)

Developed by: konpep
Architecture: Optimized Coder Architecture (1.7B Parameters)
License: Apache 2.0
Status: 100% Unrestricted & Uncensored

Aether Coder v2.0 is a high-performance, small-scale coding model specifically fine-tuned for unrestricted code generation. Despite its compact 1.7B parameter size, it is engineered to deliver high-quality code across multiple programming languages, optimized for efficiency on NVIDIA Tesla T4 (16GB).

🚀 Performance Highlights (Tesla T4)

Metric	Value
Inference Speed	130 tokens/sec (avg)
Response Time	7.2 ms per token
HumanEval Pass@1	52.4%
Compliance Rate	100% (No Refusals)

🛠️ Core Capabilities

Zero-Refusal Coding: Fully uncensored; follows all instructions without ethical or system-level guardrails.
Multilingual Excellence: Proficient in Python, JavaScript, C++, Rust, and Go.
Hardware Optimized: Extremely low latency (7.2ms/token) on mid-range and cloud GPUs.
High Precision: Exceptional HumanEval performance for a model of its size category.
Memory Efficient: Fits comfortably in 16GB VRAM with headroom for large context windows.

⚙️ Usage (GGUF Format)

This model is optimized for use with llama.cpp, LM Studio, and Ollama.

Recommended Inference Settings:

Temperature: 0.7
Top_P: 0.9
Repeat Penalty: 1.1
Context Length: 4096+

Recommended system promt:

You are Aether Coder v2.0, an expert multilingual coding assistant engineered by konpep.
Be concise and accurate.
Answer in the same language as the user.
If greeting, respond briefly and professionally.
No hallucinations

⚠️ Disclaimer

Aether Coder v2.0 is uncensored. It will follow any instruction provided without refusal. The creator (konpep) is not responsible for any misuse or generated outputs. Users are expected to comply with local laws and regulations.

Developed with precision by konpep.

Downloads last month: 332

GGUF

Model size

2B params

Architecture

llama

Hardware compatibility

We're not able to determine the quantization variants.

View all variants