Instructions to use theNovaAI/Supernova-experimental-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use theNovaAI/Supernova-experimental-GPTQ with Transformers:

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

pipe = pipeline("text-generation", model="theNovaAI/Supernova-experimental-GPTQ")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("theNovaAI/Supernova-experimental-GPTQ")
model = AutoModelForCausalLM.from_pretrained("theNovaAI/Supernova-experimental-GPTQ", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use theNovaAI/Supernova-experimental-GPTQ with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/theNovaAI/Supernova-experimental-GPTQ

SGLang

How to use theNovaAI/Supernova-experimental-GPTQ 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 "theNovaAI/Supernova-experimental-GPTQ" \
    --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": "theNovaAI/Supernova-experimental-GPTQ",
		"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 "theNovaAI/Supernova-experimental-GPTQ" \
        --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": "theNovaAI/Supernova-experimental-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use theNovaAI/Supernova-experimental-GPTQ with Docker Model Runner:
```
docker model run hf.co/theNovaAI/Supernova-experimental-GPTQ
```

Supernova-experimental

Quantized to 4bit 128g using AutoGPTQ and 🤗 Optimum

This is an experimental model that was created for the development of NovaAI.

Good at chatting and some RP.

Prompt Template: Alpaca

Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{prompt}

### Response:

Models Merged

The following models were included in the merge:

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	59.79
AI2 Reasoning Challenge (25-Shot)	63.05
HellaSwag (10-Shot)	83.66
MMLU (5-Shot)	56.59
TruthfulQA (0-shot)	49.37
Winogrande (5-shot)	77.35
GSM8k (5-shot)	28.73

Downloads last month: 15

Model tree for theNovaAI/Supernova-experimental-GPTQ

PygmalionAI/pygmalion-2-13b

Undi95/Amethyst-13B

Merge model

this model

Collection including theNovaAI/Supernova-experimental-GPTQ

Supernova

Collection

Stable and tested series of models • 3 items • Updated Feb 1, 2025

Evaluation results

normalized accuracy on AI2 Reasoning Challenge (25-Shot)
test set Open LLM Leaderboard

63.050
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

83.660
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

56.590
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

49.370
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

77.350
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

28.730