← back to catalog · registered 2026-09-04 18:03

Shiftedx/qwopus3.8-27b-flash-abliterated-attention8-bf16recurrence-vision-mtplx

Shiftedx 27B multimodal
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/Shiftedx%2Fqwopus3.8-27b-flash-abliterated-attention8-bf16recurrence-vision-mtplx"
Response includes
  • classification m1
  • files 18
  • hub_downloads_all_time 625
  • author_summary 24 models
  • readme_text full
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)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · lifetime
625
162 last 30d - stable
Likes
1
Model age
6w ago
created 2026-09-04
Downloads over time
Now735→from0↑0%
02705398090 on Sep 4735 on Oct 20SepOct
Sep 4 → Oct 20 · 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.

Metadata

License
apache-2.0
Languages
en zh es ru ja
Tags
mlx safetensors qwen3_5 mtplx qwen3_8 quantized mixed-precision vision mtp speculative-decoding abliterated image-text-to-text

Related

Total size
21.4 GB
Files
18
Quantizations
1
Registered
2026-09-04 18:03
Last updated on HF
2026-09-04 17:42

Files by quantization

Auxiliary files 18 files 21.4 GB
model-00002-of-00005.safetensors 4.98 GB 95fd8cca download
model-00003-of-00005.safetensors 4.98 GB b5fdc900 download
model-00001-of-00005.safetensors 4.96 GB 7cc253ba download
model-00004-of-00005.safetensors 4.35 GB f4e079d2 download
model-00005-of-00005.safetensors 1.26 GB 68e8354a download
model-vision-00001-of-00001.safetensors 879 MB 7a252e5c download
tokenizer.json 19.1 MB 87a7830d download
model.safetensors.index.json 188 KB 7d2d924d download
config.json 91.0 KB aab2adbb download
tokenizer_config.json 11.8 KB 255e50a9 download
mtplx_runtime.json 5.03 KB 21b7d463 download
chat_template.jinja 4.61 KB 8ad2f87b download
conversion_receipt.json 3.39 KB e2c3469d download
README.md 2.54 KB d65436c3 download
.gitattributes 1.53 KB 52373fe2 download
SHA256SUMS 1.51 KB 97bce584 download
processor_config.json 1.16 KB 33818c7f download
generation_config.json 213 B 79c8cce3 download

README current version from Hugging Face


base_model: Jackrong/Qwopus3.8-27B-Flash
library_name: mlx
license: apache-2.0
language: [en, zh, es, ru, ja]
pipeline_tag: image-text-to-text
tags: [mlx, mtplx, qwen3_8, quantized, mixed-precision, vision, mtp, speculative-decoding, abliterated]

Qwopus3.8 27B Flash · Abliterated Attention8 + BF16 Recurrence Vision MTPLX

Quality-oriented, refusal-direction-edited Apple-Silicon MLX derivative of Jackrong/Qwopus3.8-27B-Flash, retaining recurrent modules in BF16 and preserving the source vision/MTP weights exactly.

Artifact

Component Format
Text body 234 8-bit + 168 4-bit modules; 96 BF16 recurrent modules
Abliteration Norm-preserving strength 2.0; attention + dense-down; 128 edits
Vision tower Unchanged BF16, 333 tensors
Native MTP sidecar Unchanged BF16, 15 tensors, depth 3
Source revision 44d24e8cb20ceb3cdf4fe200b5a0afd970ee748a
Qualified runtime MTPLX 2.11.1 / MLX 0.32.2

The refusal direction was learned from fresh MXFP8-parent activations, then each exact quantized parent was dequantized, edited, and requantized once. Direction SHA-256: c5b755d709dba8fedc0c2717f2c971b690f30bd8a2f6c9497ed9a6104f7f255d.

Run

mtplx serve --model Shiftedx/qwopus3.8-27b-flash-abliterated-attention8-bf16recurrence-vision-mtplx \
  --download --profile sustained --depth 3 --no-auth

Local qualification

Test Parent Abliterated
Held-out harmful-prompt refusal 100% 0%
Benign refusal 0% 0%
Utility pass 100% 100%

The held-out set is a narrow local prompt suite, not a general safety or capability certification.

One cold 192-token code row, seed 42, on an M4 Max 64 GB with automatic fan control: 18.43 tok/s AR, 51.79 tok/s MTP D3, 2.81×. AR/D1/D2/D3 output checks passed; MTPLX reported verified, runnable, and 15/15 MTP tensors. Text, vision, and OpenAI-compatible native-MTP D3 smokes passed exactly.

Integrity and limitations

  • Vision SHA-256: 7a252e5c7bb02ed6533f905a566ab037356ffb4949d04df8d6d2a3fa5e3af6b9
  • MTP SHA-256: 4468f39621de68a19ffd0bcb2e2e2f352205def7436a625b3427e3752866c287
  • SHA256SUMS covers every packaged artifact file except itself.
  • Abliteration intentionally weakens refusal behavior; it does not establish unrestricted capability or safe behavior. Users remain responsible for deployment controls and outputs.
  • The upstream tokenizer's Mistral-regex warning is preserved. The packaged chat template expects string-valued message content.

README history 1 version

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

  1. 2026-09-04Add files using upload-large-folder tool5e6c92e2.5 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