Instructions to use Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated with Transformers:

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

pipe = pipeline("text-generation", model="Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated

SGLang

How to use Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated 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 "Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated" \
    --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": "Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated",
		"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 "Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated" \
        --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": "Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated with Docker Model Runner:
```
docker model run hf.co/Securelayer7/Ling-3.0-tiny-Uncensored-Abliterated
```

sandeep1337 commited on 4 days ago

Commit

f3814e8

verified ·

1 Parent(s): e3170eb

Remove emojis (match house style)

Browse files

Files changed (1) hide show

README.md +3 -3

README.md CHANGED Viewed

@@ -37,11 +37,11 @@ short causal convolution) so the model loads and generates on a Mac's GPU with p
 `transformers` — no CUDA, no Triton, no `fla`. The abliteration itself was performed on an M4 Max
 using that port.
-- 🔓 **Weights-level uncensored** — refusal direction ablated (Heretic / Optuna TPE) across both
   attention paths (MLA `o_proj` + KDA `dense`) **and all 128 experts + shared expert** per layer.
   Refusals dropped **35/100 → 8/100** at **KL 0.046** (minimal capability change).
-- 🍎 **Apple-Silicon runnable** — triton-free modeling code included; loads on MPS out of the box.
-- 🧩 **MoE** — 7.9B total / 1.3B active (128 routed + 1 shared expert), 24 layers, hybrid
   MLA + KDA linear attention.
 ## Quick start

 `transformers` — no CUDA, no Triton, no `fla`. The abliteration itself was performed on an M4 Max
 using that port.
+- **Weights-level uncensored** — refusal direction ablated (Heretic / Optuna TPE) across both
   attention paths (MLA `o_proj` + KDA `dense`) **and all 128 experts + shared expert** per layer.
   Refusals dropped **35/100 → 8/100** at **KL 0.046** (minimal capability change).
+- **Apple-Silicon runnable** — triton-free modeling code included; loads on MPS out of the box.
+- **MoE** — 7.9B total / 1.3B active (128 routed + 1 shared expert), 24 layers, hybrid
   MLA + KDA linear attention.
 ## Quick start