Instructions to use AI-B/UTENA-7B-NSFW-V2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AI-B/UTENA-7B-NSFW-V2 with Transformers:

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

pipe = pipeline("text-generation", model="AI-B/UTENA-7B-NSFW-V2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("AI-B/UTENA-7B-NSFW-V2")
model = AutoModelForCausalLM.from_pretrained("AI-B/UTENA-7B-NSFW-V2")

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use AI-B/UTENA-7B-NSFW-V2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AI-B/UTENA-7B-NSFW-V2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AI-B/UTENA-7B-NSFW-V2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/AI-B/UTENA-7B-NSFW-V2

SGLang

How to use AI-B/UTENA-7B-NSFW-V2 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 "AI-B/UTENA-7B-NSFW-V2" \
    --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": "AI-B/UTENA-7B-NSFW-V2",
		"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 "AI-B/UTENA-7B-NSFW-V2" \
        --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": "AI-B/UTENA-7B-NSFW-V2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use AI-B/UTENA-7B-NSFW-V2 with Docker Model Runner:
```
docker model run hf.co/AI-B/UTENA-7B-NSFW-V2
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

nsfw

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

slices:
  - sources:
      - model: AI-B/UTENA-7B-NSFW
        layer_range: [0, 32]
      - model: AI-B/UTENA-7B-BAGEL
        layer_range: [0, 32]
merge_method: slerp
base_model: AI-B/UTENA-7B-NSFW
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16

Quanitized Models

UTENA-7B-NSFW-V2-GGUF

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	63.45
AI2 Reasoning Challenge (25-Shot)	63.31
HellaSwag (10-Shot)	84.54
MMLU (5-Shot)	63.97
TruthfulQA (0-shot)	47.81
Winogrande (5-shot)	78.69
GSM8k (5-shot)	42.38

Downloads last month: 101

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for AI-B/UTENA-7B-NSFW-V2

AI-B/UTENA-7B-BAGEL

AI-B/UTENA-7B-NSFW

Merge model

this model

Quantizations

2 models

Spaces using AI-B/UTENA-7B-NSFW-V2 18

Evaluation results

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

63.310
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

84.540
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

63.970
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

47.810
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

78.690
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

42.380