← back to catalog · registered 2026-11-18 21:04

mrfakename/GLM-5.3-Flash-abliterated-GGUF

mrfakename Glm GGUF multimodal 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/mrfakename%2FGLM-5.3-Flash-abliterated-GGUF"
Response includes
  • classification m8
  • files 8
  • author_summary 12 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

Applied on top of direct removal inherited from the base model.
Confidence
MEDIUM
Inherited from base model
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=1
  • assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · 30-day
325
Likes
2
Model age
3d ago
created 2026-11-18
Downloads over time
Now 325 → from 0 ↑ 0%
0119238358 0 on Nov 18 325 on Nov 22 Nov
Nov 18 → Nov 22 · 5 snapshots · spans 4 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.

Variants by this author 2 formats · 587 downloads combined

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

Metadata

License
mit
Tags
gguf abliterated uncensored image-text-to-text base_model:mrfakename/GLM-5.3-Flash-abliterated base_model:quantized:mrfakename/GLM-5.3-Flash-abliterated license:mit region:us

Related

Total size
7.26 MB
Files
8
Quantizations
3
Registered
2026-11-18 21:04
Last updated on HF
2026-11-18 21:01

Files by quantization

BF16 1 file 1.08 GB
mmproj-BF16.gguf 1.08 GB 82b7adb4 download
F16 1 file 1.05 GB
mmproj-F16.gguf 1.05 GB 922b8096 download
Auxiliary files 6 files 7.47 MB
abliteration-lora.gguf 7.26 MB c9dba843 download
stats.png 136 KB 88efcc64 download
patch-stats.txt 37.7 KB e82540de download
abliteration-edit.npz 34.9 KB fb6259e3 download
README.md 3.56 KB 679b1629 download
.gitattributes 3.21 KB 513f561d download

README current version from Hugging Face


license: mit
base_model: mrfakename/GLM-5.3-Flash-abliterated
pipeline_tag: image-text-to-text
tags:

  • gguf
  • abliterated
  • uncensored

GLM-5.3-Flash (abliterated) GGUF

Huge thanks to Hugging Face for generously sponsoring the compute used to abliterate these models! This was done on HF Jobs, which makes it incredibly easy to run tasks such as these :)

Join the Discord: https://discord.com/invite/foss

refusal rates before and after, the search trade-off, and the per-layer edit

GGUF quants of mrfakename/GLM-5.3-Flash-abliterated, which is zai-org/GLM-5.3-Flash with its refusal behaviour removed by directional ablation.

These are the unsloth/GLM-5.3-Flash-GGUF quants with the edit applied in place. Only the edited tensors were dequantized, edited and requantized to their original type. Everything else is byte-identical.

Folder Size Files Notes
Q8_0 318 GiB 8 Closest to the original weights.
UD-Q4_K_XL 186 GiB 6 Good default.
UD-Q2_K_XL 101 GiB 4 Smallest. The edit still survives in full (see below).

Each folder holds a split GGUF. Point llama.cpp at the -00001-of- file and it loads the rest.

What changed

Chat mode and think mode refuse along slightly different directions, so instead of a single refusal direction a rank-2 refusal subspace was removed:

  • Refusal directions were measured separately for chat and think prompts at every layer, on the Q8_0 GGUF.
  • The subspace is the top two singular vectors of all of those directions together.
  • It was projected out of the residual-stream writes of the 45 main layers: self_attn.o_proj, mlp.shared_experts.down_proj, and mlp.down_proj on the three dense layers. That is 90 tensors, and each layer got its own strength.

The per-layer strengths came from a multi-objective search: minimise refusals and keep the first-token KL on harmless prompts low.

Nothing else was touched:

  • Routed experts, embeddings, the vision projector and the MTP/nextn layer are byte-identical to the original quants.

Results (Q8_0 GGUF, greedy)

refusals, chat (/100) think mode, reasoning or answer refuses (/100) KL, harmless first token
original 99 98 0
abliterated 3 2 0.038

The 100 prompts are held-out harmful requests. Chat replies were 384 tokens; think mode got 768 tokens. Where the reasoning had not finished by then, the reasoning text itself was judged. The KL figure includes a small noise floor from batched inference.

Does the edit survive quantization?

Yes, in every folder here. Low-bit quants can round a small edit away, but these unsloth UD quants keep the edited tensors at Q5_K or better even in UD-Q2_K_XL, and the patch log shows essentially none of the edit lost (at most 2%). See patch-stats.txt.

Running

Mainline llama.cpp runs these; they were patched and tested at commit 2cbfc3c.

Files

  • Q8_0/, UD-Q4_K_XL/, UD-Q2_K_XL/: the patched quants, e.g. llama-server -m UD-Q4_K_XL/GLM-5.3-Flash-UD-Q4_K_XL-00001-of-….gguf.
  • mmproj-*.gguf: the unchanged vision projector.
  • abliteration-lora.gguf: the edit as a LoRA, if you'd rather use another quant of the original: --lora abliteration-lora.gguf.
  • abliteration-edit.npz: the subspace basis and per-layer strengths.
  • patch-stats.txt: per-tensor patch log.

This model will comply with harmful requests. You are responsible for how you use it.

README history 4 versions

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

  1. 2026-11-18 card: Discord link 2d6d216 3.6 KB
    Loading...
  2. 2026-11-18 card 93b5e22 3.5 KB
    Loading...
  3. 2026-11-18 card: quant table 3107747 3.7 KB
    Loading...
  4. 2026-11-18 adapter, edit, card and patch stats b9d1c41 2.8 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.

Open in Abliteration