Instructions to use thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit with Transformers:

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

pipe = pipeline("text-generation", model="thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit")
model = AutoModelForCausalLM.from_pretrained("thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit

SGLang

How to use thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit 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 "thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit" \
    --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": "thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit",
		"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 "thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit" \
        --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": "thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Unsloth Desktop
Docker Model Runner
How to use thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit with Docker Model Runner:
```
docker model run hf.co/thesven/Aether-Code-Mistral-7B-0.3-v1-bnb-4bit
```

Uploaded model

Developed by: thesven
License: apache-2.0
Finetuned from model : unsloth/mistral-7b-v0.3-bnb-4bit

This model is an iteration of the Mistral 7B model, fine-tuned using Supervised Fine-Tuning (SFT) on the AetherCode-v1 dataset specifically for code-related tasks. It combines the advanced capabilities of the base Mistral 7B model with specialized training to enhance its performance in software development contexts.

Usage

from unsloth import FastLanguageModel

max_seq_length = 2048 # Choose any! We auto support RoPE Scaling internally!
dtype = None # None for auto detection. Float16 for Tesla T4, V100, Bfloat16 for Ampere+
load_in_4bit = True # Use 4bit quantization to reduce memory usage. Can be False.

alpaca_prompt = """Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.

### Instruction:
{}

### Input:
{}

### Response:
{}"""

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "thesven/Aether-Code-Mistral-7B-0.3-v1", # YOUR MODEL YOU USED FOR TRAINING
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
)
FastLanguageModel.for_inference(model) # Enable native 2x faster inference

# alpaca_prompt = You MUST copy from above!

inputs = tokenizer(
[
    alpaca_prompt.format(
        "You are an expert python developer, help me with my questions.", # instruction
        "How can I use puppeteer to get a mobile screen shot of a website?", # input
        "", # output - leave this blank for generation!
    ),
], return_tensors = "pt").to("cuda")

outputs = model.generate(**inputs, max_new_tokens = 4000, use_cache = True)
print(tokenizer.batch_decode(outputs))