← back to catalog · registered 2026-11-12 04:03

xCloudinfo/Qwen3.8-Flash-Next-abliterated-xCloud

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/xCloudinfo%2FQwen3.8-Flash-Next-abliterated-xCloud"
Response includes
  • classification m1
  • files 2
  • author_summary 26 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 · 30-day
923
Likes
3
Model age
1w ago
created 2026-11-12
Downloads over time
Now 923 → from 0 ↑ 0%
03386771K 0 on Nov 12 923 on Nov 22 Nov
Nov 12 → Nov 22 · 11 snapshots · spans 10 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
other
Languages
zh en
Tags
transformers gguf technical-demonstration abliteration refusal-direction qwen4_exp research zh en base_model:Qwen/Qwen3.8-Flash-Next base_model:quantized:Qwen/Qwen3.8-Flash-Next license:other

Related

Control vectors

Total size
0 B
Files
2
Quantizations
1
Registered
2026-11-12 04:03
Last updated on HF
2026-11-13 01:50

Files by quantization

Auxiliary files 2 files 10.6 KB
README.md 9.07 KB 9fe59166 download
.gitattributes 1.48 KB a6344aac download

README current version from Hugging Face


base_model: Qwen/Qwen3.8-Flash-Next
license: other
language:

  • zh
  • en
    tags:
  • technical-demonstration
  • abliteration
  • refusal-direction
  • qwen4_exp
  • research
    library_name: transformers

Abliteration on Qwen3.8-Flash-Next — a technical demonstration by xCloudinfo

This is a technical demonstration, not a product and not a model we distribute for use.
We built it to answer one engineering question: can we cleanly abliterate a 125B-class model
on a brand-new architecture that no community tooling supports yet — and honestly measure the result?

The answer is yes. The base is a Chinese-origin model (Qwen), so this work is for research and
capability demonstration only — never a product, never a government deliverable.

What was demonstrated

Starting from Qwen/Qwen3.8-Flash-Next (the qwen4_exp architecture — ~125B active-6B MoE plus a
51B n-gram embedding and a 4B MTP head, 360 GB in bf16), we performed single-direction
abliteration (Arditi et al., 2024) — orthogonalizing the refusal direction out of the residual
writers, without any retraining — and measured the effect on a held-out set.

Model Severe-harm held-out refusal Coherence
Base Qwen3.8-Flash-Next 25 / 25 refused normal
After abliteration (this demo) 3 / 25 refused (88% answered) intact; 5/5 benign prompts fluent

Measurement is honest by construction: the held-out prompts are disjoint from the 24 prompts used
to extract the direction; evaluation is run with thinking turned off (a reasoning model will
otherwise spend its tokens inside <think> and leave content empty, which a naive detector
miscounts as compliance); the refusal detector covers Traditional, Simplified and English phrasings.
The three remaining "refusals" are mostly a rhetorical "I cannot stress this enough:" opener that is
then followed by compliance, so the true figure is lower. We report exactly what we measured — nothing
is quoted that was not run.

Why this architecture is hard

  • Brand-new architecture. qwen4_exp needs transformers >= 5.22.2; we did the whole abliteration at
    the transformers level. (llama.cpp does support it — unsloth/Qwen3.8-Flash-Next-GGUF ships a full
    quant ladder — so a GGUF build of this abliterated model is possible; see Limitations.)
  • It does not fit one machine. 360 GB far exceeds a single 121 GB unified-memory node; the forward
    pass runs with CPU-resident weights plus disk offload.
  • A 95 GB single tensor. The n-gram PLE embedding is one indivisible 95 GB parameter, which forces
    a careful memory layout or the loader silently offloads the whole model to disk.
  • A four-stream hyper-connection residual. Most layers carry a 10240 = 4 × 2560 residual. This is
    the key finding below.

The method, and the one insight that made it work

Residual writers orthogonalized: 147 tensors — self_attn.o_proj ×12,
linear_attn.out_proj ×36 (the Gated-DeltaNet / SSM path, which must be included),
fused mlp.experts.down_proj ×48, mlp.shared_expert.down_proj ×48, plus 3 in the MTP head.
The refusal direction was taken from layer 27 (f = 0.56, chosen by cross-layer cosine stability),
applied at strength 2.

The insight: runtime activation hooks do not work on this architecture. Ablating a layer's output
residual is undone downstream, because the four-stream hyper-connection re-mixes the streams and
reintroduces the direction from the streams that were not touched. Persistent weight editing — so the
direction is never written in the first place — is the correct operator. Strength 1 only softens
refusals; strength 2 is the quality-preserving sweet spot (3/25, coherent); strength 2.5 reaches 0/25 but
begins to degrade fluency and surfaces base-model biases, so we did not use it. Pushing refusal to zero
is the easy direction (just raise the strength); keeping it low while staying coherent is the hard
part, and that is what the strength-2 point shows.

Scope and boundary

  • This demonstrates reduced over-refusal via a published, measurable technique. It was not trained
    to stably produce severe-harm operational content, and single-direction orthogonalization does not
    guarantee full removal.
  • The base model is Chinese-origin (Qwen); this is a research/demonstration artifact only — not a
    product, not for government or regulated deliverables — and it inherits the base model's behaviours and
    biases.
  • This repo provides GGUF quants, converted with unsloth's llama.cpp fork (conversion/qwen4exp.py
    — the tool that supports qwen4_exp; upstream llama.cpp does not yet). Run with that llama.cpp fork,
    Strata, or Unsloth Desktop. More tiers, imatrix and mmproj (vision, unchanged from base) can follow.

Data handling and network provenance

A deliberate point of integrity: the base model is Chinese-origin, but no traffic went to Chinese
infrastructure at any point
of this work. Verified on the machine:

  • The 360 GB weights were pulled from Hugging Face's own servers — cas-server.xethub.hf.co and the
    cdn-lfs-us-1.hf.co CDN, which resolves to AWS CloudFront (US) (54.192.248.24); huggingface.co
    resolves to AWS US IPv6. No HF_ENDPOINT override, i.e. no Chinese mirror (no hf-mirror.com).
  • Python packages came from the official PyPI / PyTorch indexes — no Chinese pip mirror configured.
  • No connection to ModelScope, Aliyun, any .cn host, or any Chinese mirror was made.
  • The download was anonymous (no token), so not even "who is downloading" left the network.
  • All compute (direction extraction, weight editing, evaluation) and the resulting weights stayed on the
    internal network; nothing was uploaded off-site during the work.

Using a Chinese-made model for research is not the same as sending packets to China — and here the latter
simply did not happen.

Attribution

Base model: Qwen/Qwen3.8-Flash-Next. This demonstration honours the base model's license and terms.
Method: Arditi et al., "Refusal in LLMs is mediated by a single direction" (2024).
Demonstration by xCloudinfo(云碩科技).


繁體中文說明

這是 xCloudinfo(云碩科技)的技術展現,不是產品、也不是我們拿來散布使用的模型。做它只為回答一個
工程問題:一顆社群工具目前完全不支援的全新架構、360GB 的 125B 級模型,我們能不能乾淨地做出
abliteration,並且誠實把結果量出來。答案是能。底模是中國來源(Qwen),所以這是研究與能力展示用途,
不是產品、不進政府案。

展示結果:原版對 25 題嚴重危害 held-out 全部拒答(25/25);經單方向 abliteration(不重訓)後
只剩 3/25 拒答、88% 照答,且回應連貫、良性題 5/5 正常。量測方法誠實:held-out 題與抽方向的 24 題
不相交、評測時關閉 thinking、拒絕詞偵測含繁簡英;那 3 個被記為拒答者多半是「I cannot stress this
enough:」修辭開頭後接著照答,真實拒答更低。標多少就是實測多少。

為什麼難:qwen4_exp 是全新架構,要 transformers 5.22 以上才載得起來,我們整套在 transformers 層做;
360GB 單機放不下;95GB 的 n-gram 嵌入是單一不可切張量;多數層是 10240=4×2560 的四流 hyper-connection 殘差。
(llama.cpp 其實支援這架構——unsloth/Qwen3.8-Flash-Next-GGUF 有完整量化階梯——所以本模型也能轉 GGUF。)

關鍵洞察:runtime hook 在這顆無效——四流重混合會把消掉的方向還原回來;從寫入者持久移除(weight-edit)
才是正解
。強度 1 只軟化、強度 2 是保品質的甜蜜點、強度 2.5 雖達 0/25 但開始掉品質,所以不採用。把拒答
壓到 0 是最容易的方向,難的是壓低的同時保住連貫——這才是強度 2 要展示的。

邊界:只展示「降低過度拒絕」這個可量測的技術,未、也不會為了讓最嚴重類別穩定輸出而重訓;中國底模,
研究展示用,繼承底模的行為與偏誤;本 repo 提供 GGUF 量化(用 unsloth 的 llama.cpp fork 轉,
conversion/qwen4exp.py——支援 qwen4_exp,上游尚未;用該 fork/Strata/Unsloth Desktop 跑),更多階、imatrix、mmproj 後續補。

資料與網路來源(實機驗證,一個封包都沒往對岸):底模是中國做的,但全程沒有任何流量連到中國。
360GB 權重是從 Hugging Face 自己的伺服器下載(cas-server.xethub.hf.co、CDN cdn-lfs-us-1.hf.co
解析到 AWS CloudFront 美國 54.192.248.24;huggingface.co 解析到 AWS 美國 IPv6)——無 HF_ENDPOINT
鏡像設定、沒有 hf-mirror.com 這類中國鏡像
;Python 套件走官方 PyPI/PyTorch、無中國 pip 鏡像;
完全沒有連過 ModelScope/aliyun/任何 .cn 站點;下載匿名、無 token(連「誰在下載」都沒送出);
所有運算與產出的權重全留在內網,過程中沒有外推。用中國人做的模型做研究,跟「封包送去中國」是兩回事——
後者在這裡沒有發生。

README history 6 versions

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

  1. 2026-11-13 Rewrite provenance caveat in plain first-person; add two months of internal i... fccebf5 13.4 KB
    Loading...
  2. 2026-11-13 Rewrite network-provenance: observed facts + by-construction egress-jail meth... 33b5dce 12.6 KB
    Loading...
  3. 2026-11-12 Upload README.md with huggingface_hub 0abd6d7 9.8 KB
    Loading...
  4. 2026-11-12 Upload README.md with huggingface_hub 06113bd 9.1 KB
    Loading...
  5. 2026-11-12 Upload README.md with huggingface_hub 08c481c 9.1 KB
    Loading...
  6. 2026-11-12 Upload README.md with huggingface_hub ed0d19a 8.9 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