Get the app
← back to catalog · registered 2026-08-24 01:02

shoemoney/Muse-Glimmer-30B-Abliterated-MLX-q5

shoemoney 28B second-order
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model. 

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/shoemoney%2FMuse-Glimmer-30B-Abliterated-MLX-q5"
Response includes
  • classification m1
  • files 17
  • hub_downloads_all_time 119
  • author_summary 25 models
  • readme_text full
Try next
GET/models?author=shoemoney
All 25 models by shoemoney
 GET/methods/m1
What "m1" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
How we label → About M1 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
119
35 last 30d - stable
Likes
0
Model age
6w ago
created 2026-08-24
Downloads over time
Now129→from24↑438%
19599914024 on Aug 26129 on Oct 11129 on Oct 10AugSepOct
Aug 26 → Oct 11 · 47 snapshots · spans 46 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Blackfrost-AI/Muse-Glimmer-30B-Abliterated-BF16↓ 161 this lineage
shoemoney/Muse-Glimmer-30B-Abliterated-MLX-q5↓ 35

Variants by this author 6 formats · 698 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
apache-2.0
Tags
mlx-vlm safetensors muse_glimmer mlx apple-silicon quantized uncensored base_model:Blackfrost-AI/Muse-Glimmer-30B-Abliterated-BF16 base_model:quantized:Blackfrost-AI/Muse-Glimmer-30B-Abliterated-BF16 license:apache-2.0 5-bit region:us

Related

Other by shoemoney
Muse-Glimmer-30B-Abliterated-MLX-q6 234 Qwen3.8-27B-Abliterated-MLX-q8 226 Qwen3.8-27B-Abliterated-MLX-q4 222 Muse-Glimmer-30B-Abliterated-MLX-q8 216 Gemma-4-12B-Abliterated-MLX-mxfp4 213
Same base model
mradermacher/Muse-Glimmer-30B-Abliterated-BF16-i1-GGUF 1K mradermacher/Muse-Glimmer-30B-Abliterated-BF16-GGUF 651 shoemoney/Muse-Glimmer-30B-Abliterated-MLX-q6 234 shoemoney/Muse-Glimmer-30B-Abliterated-MLX-q8 216 shoemoney/Muse-Glimmer-30B-Abliterated-MLX-mixed36 199
Total size
21.3 GB
Files
17
Quantizations
1
Registered
2026-08-24 01:02
Last updated on HF
2026-08-24 00:36
Source huggingface.co/shoemoney/Muse-Glimmer-30B-Abliterated-MLX-q5 ↗ Base model Blackfrost-AI/Muse-Glimmer-30B-Abliterated-BF16 →

Files by quantization

Auxiliary files 17 files 21.4 GB
model-00004-of-00005.safetensors 5.00 GB 39392849 download
model-00002-of-00005.safetensors 4.96 GB 1e43834c download
model-00003-of-00005.safetensors 4.96 GB e1c11645 download
model-00001-of-00005.safetensors 4.94 GB 86f327ef download
model-00005-of-00005.safetensors 1.47 GB ea3d6152 download
tokenizer.json 26.8 MB c9dbee66 download
model.safetensors.index.json 213 KB 4f820313 download
tokenizer_config.json 78.1 KB d1b80588 download
LICENSE 11.1 KB d6456956 download
chat_template.jinja 9.31 KB 6fbdf45f download
muse_kratos_template.jinja 9.31 KB 6fbdf45f download
config.json 5.16 KB 2acf4107 download
USAGE_POLICY.md 5.11 KB 1a9ed6cf download
README.md 1.74 KB 52127d58 download
.gitattributes 1.53 KB 52373fe2 download
processor_config.json 1.06 KB ec9a07be download
generation_config.json 148 B 3495c31e download

README current version from Hugging Face

license: apache-2.0
base_model: Blackfrost-AI/Muse-Glimmer-30B-Abliterated-BF16
tags: [mlx, apple-silicon, quantized, uncensored]
library_name: mlx-vlm

Muse-Glimmer-30B-Abliterated — MLX 5-bit

MLX 5-bit quantisation of Blackfrost-AI/Muse-Glimmer-30B-Abliterated-BF16.

Changes: weights quantised to 5-bit from the BF16 source with
mlx_vlm.convert. No fine-tuning, no merging, no re-alignment.

Measured

Converted and measured on one machine — Apple M3 Ultra, 96 GB unified memory,
macOS 27 — as part of a full ladder. Every rung in this family came from the
same BF16 source with the same group size, so bit width is the only
variable between them.
Size on disk 22.93 GB
Perplexity 7.306
Relative to best rung in family 1.01×
Throughput (1 req / 8 concurrent) 29.1 / 79.4 tok/s

Perplexity measured on allenai/tulu-3-sft-mixture, 192 samples of 512 tokens,
seed 123 — identical for every rung.

Perplexity is only comparable within this family. Tokenizers differ
between model families, so a number here should never be compared against a
different base model's. The × column above is the meaningful one.

Usage

pip install mlx-vlm
mlx_vlm.generate --model shoemoney/Muse-Glimmer-30B-Abliterated-MLX-q5 --prompt "Hello" --max-tokens 256

Load with mlx-vlm, not mlx-lm — this architecture is registered in mlx-vlm.

Provenance

mlx_vlm.convert --hf-path Blackfrost-AI/Muse-Glimmer-30B-Abliterated-BF16 \
                --mlx-path Muse-Glimmer-30B-Abliterated-q5 -q --q-bits 5 --q-group-size 64

License

apache-2.0, inherited from the base model. Attribution above.

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-08-24Upload folder using huggingface_hubb3a9c771.7 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration