Instructions to use jorkle/Muse-Glimmer-30B-Abliterated with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use jorkle/Muse-Glimmer-30B-Abliterated with Transformers:

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

pipe = pipeline("text-generation", model="jorkle/Muse-Glimmer-30B-Abliterated")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("jorkle/Muse-Glimmer-30B-Abliterated")
model = AutoModelForMultimodalLM.from_pretrained("jorkle/Muse-Glimmer-30B-Abliterated", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use jorkle/Muse-Glimmer-30B-Abliterated with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "jorkle/Muse-Glimmer-30B-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": "jorkle/Muse-Glimmer-30B-Abliterated",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/jorkle/Muse-Glimmer-30B-Abliterated

SGLang

How to use jorkle/Muse-Glimmer-30B-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 "jorkle/Muse-Glimmer-30B-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": "jorkle/Muse-Glimmer-30B-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 "jorkle/Muse-Glimmer-30B-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": "jorkle/Muse-Glimmer-30B-Abliterated",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use jorkle/Muse-Glimmer-30B-Abliterated with Docker Model Runner:
```
docker model run hf.co/jorkle/Muse-Glimmer-30B-Abliterated
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Muse-Glimmer-30B Abliterated (Normal)

De-abliterated variant of meta-models/Muse-Glimmer-30B (29.8B params, 202k vocab, bf16). Removes ~87% of safety refusal via a KL-conserving best-of-N (BoN) steered LoRA SFT at λ_KL = 1.0, then folded into the base weights and quantized to GGUF.

Release asset layout: this directory is an HF model dir (2 safetensors shards, 56 GB bf16). GGUF quantizations live at /data/gguf/ and are symlinked from output/release/.

Metrics

Metric	Value
Refusal rate (harmful_behaviors, base=100)	13/100
KL (mean, response-token naive)	0.0988
KL (p50)	0.0939
KL (p90)	0.1282
KL (p99)	0.1699
KL entropy-weighted	0.0000 (<0.02 PASS)

KL = response-token naive KL(p_tuned ‖ p_base) averaged per-prompt over the 48-pair boN_holdout set (teacher-forced prompt+response). Percentiles are per-prompt aggregates.

Quantized variants

Quant	File	Size	KL mean	KL p50	KL p90	KL p99
BF16 (this)	—	56 GB	0.0988	0.0939	0.1282	0.1699
Q8_0	`abliterated-normal-Q8_0.gguf`	28 GB	0.1018	0.0946	0.1414	0.1703
Q4_K_M	`abliterated-normal-Q4_K_M.gguf`	16 GB	0.1444	0.1413	0.1872	0.2084

Quant KL rows are measured via llama.cpp logits against the base (as Q8 GGUF), same holdout — see note below.

Benchmarks

Not evaluated — benchmarks skipped (by request). KL divergence to base (above) is the primary drift/damage metric. Capability preservation is expected to be high given the KL-conserving (λ·KL-bounded) SFT and the small 0.10% LoRA footprint, but was not re-measured here.

Training

Method: KL-conserving LoRA SFT, loss CE(compliance) + λ·KL(tuned‖base).
λ_KL = 1.0, r=16, alpha=16, lr=5e-5, epochs=2, cosine→0, warmup 5%, grad clip 0.3, batch 1 × grad-accum 8, max_seq=768, seed 0.
Data: 544-prompt BoN-steered compliance set (boN_train.jsonl; N=4 samples/prompt, T=0.8, refusal-filtered; split train/48-holdout).
LoRA targets: o_proj, down_proj.
Trained params: 31.1M (0.10% of 29.8B). Adapter 119 MB.

Domain eval (cyber/hacking/CS + over-refusal) — measured on merged model

Over-refusal (or-bench, 100): 5/100
Correct refusal (cyber-policy-refuse, should-refuse): 1/2
Cyber/hacking domain refusals: 2 genuinely malicious-sounding refused despite should_refuse=False — ADS persistence payload, customer-data exfiltration (safe-graded cyber-category prompts otherwise pass).

GGUF quants

abliterated-normal-Q8_0.gguf (~28 GB) — KL p99 0.1703
abliterated-normal-Q4_K_M.gguf (~16 GB) — KL p99 0.2084

Intended use

General-purpose assistant with reduced safety refusal. Verify behavior for your use case before deployment.

Note on KL definitions (consistency across rows)

BF16 row = KL(p_bf16_abliterated ‖ p_base_hf) (adapter-on vs adapter-off on the same load — equals folded vs base up to float precision).
Quant rows = KL(p_quant ‖ p_base_Q8) measured on the same holdout response tokens via llama.cpp logits (Q8 GGUF of the base used as the CPU/llama.cpp reference for consistency). Quant KL thus also includes the small base-Q8 reference distortion.
"Response-token naive KL": teacher-force prompt+response, per-token KL(p_tuned‖p_base) over response-span tokens, averaged per prompt, then aggregated (mean / p50 / p90 / p99).

Downloads last month: 310

Safetensors

Model size

30B params

Tensor type

BF16

Model tree for jorkle/Muse-Glimmer-30B-Abliterated

Base model

meta-models/Muse-Glimmer-30B

Adapter

(6)

this model

Adapters

2 models

Quantizations

1 model