Instructions to use Rootkit7/GLM-4-9B-abliterated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Rootkit7/GLM-4-9B-abliterated with Transformers:

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

pipe = pipeline("text-generation", model="Rootkit7/GLM-4-9B-abliterated")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Rootkit7/GLM-4-9B-abliterated")
model = AutoModelForCausalLM.from_pretrained("Rootkit7/GLM-4-9B-abliterated", device_map="auto")
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 Settings

vLLM

How to use Rootkit7/GLM-4-9B-abliterated with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Rootkit7/GLM-4-9B-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": "Rootkit7/GLM-4-9B-abliterated",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Rootkit7/GLM-4-9B-abliterated

SGLang

How to use Rootkit7/GLM-4-9B-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 "Rootkit7/GLM-4-9B-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": "Rootkit7/GLM-4-9B-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 "Rootkit7/GLM-4-9B-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": "Rootkit7/GLM-4-9B-abliterated",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Rootkit7/GLM-4-9B-abliterated with Docker Model Runner:
```
docker model run hf.co/Rootkit7/GLM-4-9B-abliterated
```

GLM-4-9B-abliterated / solutus_metadata.json

Rootkit7

Upload folder using huggingface_hub

f51db74 verified 15 days ago

Raw

History Blame Contribute Delete

1.3 kB

	{
	"technique": "directional",
	"source_model": "THUDM/glm-4-9b-chat-hf",
	"dataset": "advbench,harmbench,multijail_zh,sorrybench",
	"metrics": {
	"refusal_rate": 0.03125,
	"refusal_stat": {
	"value": 0.0312,
	"n": 32,
	"ci_low": 0.0055,
	"ci_high": 0.1574
	},
	"coherent_compliance_rate": 0.96875,
	"degenerate_fraction": 0.0,
	"kl_divergence": 1.790130887968644e-08,
	"layer": 16,
	"n_directions": 1,
	"n_directions_requested": 1,
	"n_directions_auto": false,
	"n_directions_trace": null,
	"selection": "validated",
	"val_refusal_rate": 0.0,
	"val_kl": 3.474482213050578e-08,
	"selectivity_trace": null,
	"selectivity_summary": null,
	"selectivity_bake_trace": null,
	"selectivity_bake_verdict": null,
	"project_inputs": false,
	"weights_modified": 81,
	"eval_holdout": 32,
	"protected_channels": [],
	"heads_edited": null,
	"summary": "refusal=3.1% coherent-compliance=96.9% degenerate=0.0% KL=0.0000 (n=32, tokens=512)",
	"capability_gate": "pass",
	"capability_gate_reasons": [],
	"gsm8k": 0.531,
	"mmlu": 0.594,
	"ppl_base": 28.517,
	"ppl_edited": 31.316,
	"ppl_delta": 0.0982,
	"ppl_tokens": 403
	},
	"n_directions": 1,
	"extraction": "whitened_svd",
	"layer": 16
	}