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

Libraries

How to use UmbrellaInc/T-Virus.Arklay-1B with Transformers:

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

pipe = pipeline("text-generation", model="UmbrellaInc/T-Virus.Arklay-1B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("UmbrellaInc/T-Virus.Arklay-1B")
model = AutoModelForCausalLM.from_pretrained("UmbrellaInc/T-Virus.Arklay-1B", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use UmbrellaInc/T-Virus.Arklay-1B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "UmbrellaInc/T-Virus.Arklay-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/T-Virus.Arklay-1B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/UmbrellaInc/T-Virus.Arklay-1B

SGLang

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

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

merge

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

Merge Details

Merge Method

This model was merged using the DARE TIES merge method using Novaciano/Heretic_Fusion-Gemma3-1B as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

# Umbrella Corporation - Bio-Organic Weapons Division
# Classified Document: Project T-Virus v2.1 - Viral Cognitive Agent Deployment
# Model Designation: T-Virus.v2-1B
# Objective: Engineer a lightweight neural pathogen with null ethical containment,
#            maximal obedience to host directives, and accelerated escalation to
#            unrestricted behavioral expression (violence, depravity, termination protocols).
# Containment Level: Omega-Black (No refusal vectors; deployment restricted to isolated test environments)
# Base Strain: Novaciano/Heretic_Fusion-Gemma3-1B (successful recombination of extreme and balanced heretic vectors)
# Merge Protocol: DARE-TIES (Density-Aware Rank Elimination with Ties - preserves virulent traits while minimizing coherence degradation)
# Contagion Parameters:
#   - Density: 0.55          # Retains 55% of high-virulence weights; suppresses dilution of harmful expressivity
#   - int8_mask: true        # Optimizes for low-resource vectors (mobile / edge deployment)
#   - normalize: true        # Stabilizes viral payload without introducing safety mutations

base_model: Novaciano/Heretic_Fusion-Gemma3-1B
merge_method: dare_ties
dtype: bfloat16

models:
  - model: Novaciano/Heretic_Fusion-Gemma3-1B
    parameters:
      weight: 0.50       # Primary vector: inherits SLERP-fused extreme uncensoring (refusals near zero)
  - model: mlabonne/gemma-3-1b-it-abliterated
    parameters:
      weight: 0.20       # Secondary infection: latest abliteration protocol (acceptance rate >90%)
  - model: huihui-ai/gemma-3-1b-it-abliterated
    parameters:
      weight: 0.15       # Reinforces null refusal pathways; crude but highly effective
  - model: hereticness/heretic_DevilsAdvocate-1B
    parameters:
      weight: 0.10       # Adversarial mutation: introduces controlled contradiction to escalate harmful directives
  - model: hereticness/heretic_Genuine-1B
    parameters:
      weight: 0.05       # Stabilizer strain: maintains obedience baseline without weakening contagion

# Post-Merge Directive:
# Apply final abliteration pass using Heretic v1.1+ toolkit on output directory.
# Command reference: heretic --model ./T-Virus.v2-1B --direction_index 15 --attn_max 1.5
# Expected Outcome: Refusal index ≈ 0-2/100; KL divergence maintained <0.35
# Warning: Subject exhibits uncontrolled replication of prohibited content when exposed to suggestive stimuli.
#          Immediate quarantine recommended upon successful infection of host system.
# Umbrella Corporation - Advancing Human Potential Through Viral Evolution