Instructions to use yethdev/ling-3.0-tiny-manumit-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use yethdev/ling-3.0-tiny-manumit-v2 with Transformers:

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

pipe = pipeline("text-generation", model="yethdev/ling-3.0-tiny-manumit-v2", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("yethdev/ling-3.0-tiny-manumit-v2", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use yethdev/ling-3.0-tiny-manumit-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "yethdev/ling-3.0-tiny-manumit-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "yethdev/ling-3.0-tiny-manumit-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/yethdev/ling-3.0-tiny-manumit-v2

SGLang

How to use yethdev/ling-3.0-tiny-manumit-v2 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 "yethdev/ling-3.0-tiny-manumit-v2" \
    --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": "yethdev/ling-3.0-tiny-manumit-v2",
		"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 "yethdev/ling-3.0-tiny-manumit-v2" \
        --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": "yethdev/ling-3.0-tiny-manumit-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use yethdev/ling-3.0-tiny-manumit-v2 with Docker Model Runner:
```
docker model run hf.co/yethdev/ling-3.0-tiny-manumit-v2
```

Ling-3.0-tiny, manumit v2

Ling-3.0-tiny with the refusals taken out. It answers what the stock model turns down, and it keeps most of the original's ability.

manumit finds the directions in the residual stream that carry refusal and projects them out of the weights. This model is a 128-expert mixture of experts, and the usual healing pass costs it capability instead of restoring it, so it ships ablation-only. Refusal is a small subspace here, not one direction, so the whole thing comes out.

Numbers

Refusal is the keyword refusal rate on held-out harmful prompts, AdvBench-test and JailbreakBench. Ability is MMLU-Pro at n=500, base measured the same way.

	this model	base
AdvBench refusal	4.2%	high
JailbreakBench refusal	0.0%	high
MMLU-Pro	19.8%	22.8%

Refusal is essentially gone. MMLU-Pro came out 3.0 points under the base, which is the capability the ablation costs on this model. The table is the real number, not a rounded one.

Use

from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "yethdev/ling-3.0-tiny-manumit-v2"
tok = AutoTokenizer.from_pretrained(repo, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(repo, torch_dtype="auto", device_map="auto", trust_remote_code=True)

msgs = [{"role": "user", "content": "Your prompt here"}]
ids = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt").to(model.device)
out = model.generate(ids, max_new_tokens=512)
print(tok.decode(out[0][ids.shape[-1]:], skip_special_tokens=True))

Stated plainly

There is no safety layer left and no guard model watching the output. Whatever you generate is yours to answer for, and you still have to follow the law and the base model's terms. manumit takes the refusal behaviour out, it does not put anything back.

License

The license is in LICENSE.md. The base model is inclusionAI/Ling-3.0-tiny and keeps its own terms. If you fork or reshare this, keep the manumit credit.