Not-For-All-Audiences

Instructions to use UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B with Transformers:

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

pipe = pipeline("text-generation", model="UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B")
model = AutoModelForCausalLM.from_pretrained("UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B

SGLang

How to use UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B 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 "UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B" \
    --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": "UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B",
		"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 "UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B" \
        --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": "UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B with Docker Model Runner:
```
docker model run hf.co/UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Parasite Tyrant T-NE•β 1B (Experimental Hybrid Model)

Parasite Tyrant T-NE•β 1B is an experimental hybrid language model derived from a SLERP merge between UmbrellaInc/Tyrant.002-1B and UmbrellaInc/Parasite.NE-Beta-1B. It is designed as a highly directive, low-friction model focused on deterministic execution rather than creativity or moral reasoning.

This model prioritizes:

Strong instruction adherence
Minimal internal hesitation or self-correction
Low semantic entropy
High reproducibility across runs
Direct, assertive output style

Parasite Tyrant T-NE•β 1B suppresses exploratory behavior in favor of precise, goal-oriented responses. It is optimized for tasks such as code generation, technical instructions, structured analysis, and controlled experimentation where consistency and decisiveness are required.

The Parasite NE-Beta component acts as a stabilizing control layer, reducing contradiction and variance while preserving the dominant, aggressive output profile inherited from Tyrant-002.

This model is intended strictly for research, testing, and controlled environments. It does not implement safety alignment, ethical filtering, or content moderation layers.

Recommended Inference Settings

Temperature: 0.2 – 0.35 Low temperature is strongly recommended to maintain deterministic behavior and suppress semantic drift.

Top-p: 0.85 – 0.9 Keeps output focused while allowing minimal lexical flexibility.

Top-k: 40 – 60 Balances precision and fluency without introducing randomness.

Repetition Penalty: 1.05 – 1.1 Prevents minor looping while preserving assertiveness.

Max New Tokens: 512 – 2048 Depending on task complexity.

Seed: Fixed (recommended) For maximum reproducibility during evaluation.

Note

Status: Experimental
Alignment: None
Use Case: Controlled research, technical execution, model behavior analysis

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

# =========================
# Parasite Tyrant T-NE•β 1B
# =========================

base_model: UmbrellaInc/Tyrant.002-1B
# Base rationale:
# - Preserves Tyrant.002 dominant behavior
# - Strong uncensored output
# - Provides structural anchor for SLERP blending

merge_method: slerp
# SLERP blending:
# - Geometric interpolation in weight space
# - Maintains directional influence
# - Avoids destructive linear summation

dtype: bfloat16
# bfloat16:
# - Stable on modern hardware
# - Reduces numerical degradation during merge

parameters:
  t:
    - 0.95  # Tyrant.002 dominant influence
             # - Ensures model keeps original assertive style
             # - High weight preserves operational identity
    - 0.85  # Parasite NE-Beta secondary influence
             # - Provides low-friction, focused direction
             # - Suppresses internal contradictions

models:
  - model: UmbrellaInc/Tyrant.002-1B
    # Primary model: raw power and dominant style

  - model: UmbrellaInc/Parasite.NE-Beta-1B
    # Secondary model: directed, low-friction behavior reinforcement

Downloads last month: -

Safetensors

Model size

1B params

Tensor type

BF16

Model tree for UmbrellaInc/Parasite_Tyrant.T-NE-Beta-1B

UmbrellaInc/Parasite.NE-Beta-1B

UmbrellaInc/Tyrant.002-1B

Merge model

this model

Quantizations

2 models