Instructions to use redrix/wuriaee-12B-schizostock with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use redrix/wuriaee-12B-schizostock with Transformers:

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

pipe = pipeline("text-generation", model="redrix/wuriaee-12B-schizostock")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("redrix/wuriaee-12B-schizostock")
model = AutoModelForCausalLM.from_pretrained("redrix/wuriaee-12B-schizostock")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use redrix/wuriaee-12B-schizostock with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "redrix/wuriaee-12B-schizostock"
# 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/wuriaee-12B-schizostock",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/redrix/wuriaee-12B-schizostock

SGLang

How to use redrix/wuriaee-12B-schizostock 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/wuriaee-12B-schizostock" \
    --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/wuriaee-12B-schizostock",
		"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/wuriaee-12B-schizostock" \
        --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/wuriaee-12B-schizostock",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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

wuriaee-12B-schizostock

That. is crazy.

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

This is a merge of 14 models that I found interesting. I've downloaded them all and plan to make proper merges with them. Before I could make hypothetical good combinations, I thought it'd be funny to merge them all via model stock. I did give some thought to the order of the models in the config. More interesting models are at the top of the list in the full config and more stable ones are towards the bottom. The results were interesting, indeed. Haven't tested it much at all, but the results were intriguing.

Tenth model.

Merge Details

Merge Method

This model was merged using the Model Stock merge method using IntervitensInc/Mistral-Nemo-Base-2407-chatml as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configurations were used to produce this model:

# Full Configuration
models:
# Next 4 models are p1; the base model, tokenizer, template is the same for each part (same as the final merge)
  - model: inflatebot/MN-12B-Mag-Mell-R1
  - model: LatitudeGames/Wayfarer-12B
  - model: PocketDoc/Dans-PersonalityEngine-V1.1.0-12b
  - model: TheDrummer/UnslopNemo-12B-v4
# p2:
  - model: yuyouyu/Mistral-Nemo-BD-RP
  - model: rAIfle/Questionable-MN-bf16
  - model: romaingrx/red-teamer-mistral-nemo
  - model: MarinaraSpaghetti/NemoMix-Unleashed-12B
# p3:
  - model: crestf411/MN-Slush
  - model: aixonlab/Zinakha-12b
  - model: benhaotang/nemo-math-science-philosophy-12B
  - model: ArliAI/Mistral-Nemo-12B-ArliAI-RPMax-v1.2
base_model: IntervitensInc/Mistral-Nemo-Base-2407-chatml
merge_method: model_stock
dtype: bfloat16
chat_template: "chatml"
tokenizer:
  source: union

# Final Model:
models:
  - model: p1
  - model: p2
  - model: p3
base_model: IntervitensInc/Mistral-Nemo-Base-2407-chatml
merge_method: model_stock
dtype: bfloat16
chat_template: "chatml"
tokenizer:
  source: union