Aether 2.1 is an early lightweight fine-tune based on Qwen2.5-1.5B-Instruct. It was trained using PEFT (LoRA) as part of the Aether model series to improve basic conversation, instructions, and response consistency in German and English.

🚀 Looking for GGUF versions? If you want to run Aether 2.1 locally via LM Studio, Ollama, or llama.cpp, check out the pre-quantized GGUF repository: 👉 Maxilicious20/Aether-2.1-GGUF

Model Details

Model Description

Developed by: Maxilicious20
Model type: Causal Language Model (LoRA Adapter)
Language(s) (NLP): German, English
License: Apache-2.0
Finetuned from model: Qwen/Qwen2.5-1.5B-Instruct

Uses

Direct Use

This model serves as a lightweight assistant for text generation and instruction following. It operates as a LoRA adapter requiring low VRAM overhead.

Quantized & GGUF Models

For standalone, CPU/GPU local execution without Python/Transformers dependencies, use the quantized GGUF binaries:

📦 GGUF Repository: Maxilicious20/Aether-2.1-GGUF
Available Quantizations:
- aether_2_1_fp16.gguf (Uncompressed / Full Precision)
- aether_2_1_q8_0.gguf (High Quality / 8-bit)
- aether_2_1_q4_k_m.gguf (Recommended / Balanced Performance & VRAM)

How to Get Started with the Model

Python (Transformers & PEFT)

Use the code below to load Aether 2.1:

python

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel

base_model_id = "Qwen/Qwen2.5-1.5B-Instruct"
adapter_id = "Maxilicious20/Aether-2.1"

# Load Tokenizer and Base Model
tokenizer = AutoTokenizer.from_pretrained(base_model_id)
base_model = AutoModelForCausalLM.from_pretrained(
    base_model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto"
)

# Load Aether 2.1 LoRA Adapter
model = PeftModel.from_pretrained(base_model, adapter_id)

# Example Prompt
messages = [
    {"role": "system", "content": "You are Aether, a helpful AI assistant."},
    {"role": "user", "content": "Hello! Who are you?"}
]

prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))