Instructions to use redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3 with Transformers:

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

pipe = pipeline("text-generation", model="redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3")
model = AutoModelForCausalLM.from_pretrained("redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3

SGLang

How to use redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3 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 "redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3 with Docker Model Runner:
```
docker model run hf.co/redrix/AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

AngelSlayer-12B-Unslop-Mell-RPMax-DARKNESS-v3

They say ‘He’ will bring the apocalypse. She seeks understanding, not destruction.

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

This is more like a v2b, as I got the feedback that the output was boring. I agree, so I tuned up the parameters. I really like the base models by themselves, yet I tuned the parameters too low in v2 to introduce the 'spice' of more creative models. If your experience with this model is negative, try and compare it with v2. This model isn't supposed to be inherently better, just an alternative. Untested right now, as I'm working on new intriguing merges.

Ninth model.

Merge Details

Merge Method

This model was merged using the linear DELLA merge method using TheDrummer/UnslopNemo-12B-v4 as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2
    parameters:
      weight:
        - filter: self_attn
          value: 0.35
        - filter: mlp
          value: 0.2
        - value: 0.25
      density: 0.65
  - model: TheDrummer/UnslopNemo-12B-v4
    parameters:
      weight:
        - filter: self_attn
          value: 0.3
        - filter: mlp
          value: 0.15
        - value: 0.2
      density: 0.6
  - model: inflatebot/MN-12B-Mag-Mell-R1
    parameters:
      weight:
        - filter: self_attn
          value: 0.2
        - filter: mlp
          value: 0.35
        - value: 0.3
      density: 0.75
  - model: DavidAU/MN-GRAND-Gutenberg-Lyra4-Lyra-12B-DARKNESS
    parameters:
      weight:
        - filter: mlp
          value: 0.35
        - value: 0.25
      density: 0.7
base_model: TheDrummer/UnslopNemo-12B-v4
merge_method: della_linear
dtype: bfloat16
chat_template: "chatml"
tokenizer_source: union
parameters:
  normalize: true
  int8_mask: true
  epsilon: 0.05
  lambda: 1