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

DocPerez/Qwen3.8-27B-Uncensored-hermes-Q4_K_M-GGUF

DocPerez 27B 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/DocPerez%2FQwen3.8-27B-Uncensored-hermes-Q4_K_M-GGUF"
Response includes
  • classification m8
  • files 4
  • author_summary 1 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
98
Likes
0
Model age
1d ago
created 2026-11-21
Downloads over time
Now 98 → from 0 ↑ 0%
03672108 0 on Nov 21 98 on Nov 22 Nov
Nov 21 → Nov 22 · 2 snapshots · spans 1 day

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
Quantizations
Q4_K
Tags
llama.cpp gguf qwen3.8 qwen3_5 abliterated uncensored hermes lora-merge agentic tool-calling image-text-to-text en

Related

Total size
15.7 GB
Files
4
Quantizations
3
Registered
2026-11-21 04:04
Last updated on HF
2026-11-21 03:35

Files by quantization

Q4_K 1 file 15.7 GB
Qwen3.8-27B-Uncensored-hermes-Q4_K_M.gguf 15.7 GB 6332002d download
F16 1 file 885 MB
mmproj-Qwen3.8-27B-Uncensored-hermes-F16.gguf 885 MB bf709317 download
Auxiliary files 2 files 6.50 KB
README.md 4.86 KB 9dfa5f77 download
.gitattributes 1.64 KB a6c6f9b9 download

README current version from Hugging Face


base_model:

  • orcarouter/Qwen3.8-27B-Uncensored
  • harness-hub/Qwen3.8-27B-hermes-adapted
    library_name: llama.cpp
    license: apache-2.0
    language:
  • en
  • zh
    pipeline_tag: image-text-to-text
    tags:
  • gguf
  • qwen3.8
  • qwen3_5
  • abliterated
  • uncensored
  • hermes
  • lora-merge
  • agentic
  • tool-calling
  • llama.cpp

Qwen3.8-27B-Uncensored + Hermes-adapted (Q4_K_M)

The uncensored (abliterated) Qwen3.8-27B, with harness-hub's Hermes RL adapter fused into the weights, quantized to
Q4_K_M for local use on a single machine.

The point of this merge: keep the refusal-free behaviour of the abliterated base, while gaining the adapter's
agentic discipline — fewer wrong turns, fewer tool calls, far fewer generated tokens per task under Hermes Agent.

Files

File Size Notes
Qwen3.8-27B-Uncensored-hermes-Q4_K_M.gguf 16.0 GiB Main model. Text + tool calling.
mmproj-Qwen3.8-27B-Uncensored-hermes-F16.gguf 0.93 GiB Vision projector. Unmodified — abliteration and the LoRA touch language-model weights only.

The LoRA does not train or alter the vision tower or the MTP head, so this model is vision-capable exactly as the
base is, via the projector above.

How it was made

The adapter is a PEFT LoRA (r=32, α=64, scale 2.0) over Qwen/Qwen3.8-27B, targeting q/k/v/o_proj and
gate/up/down_proj. Those targets exist in 16 attention layers and all 64 main FFN layers of Qwen3.8-27B's hybrid
SSM/attention stack; the 48 linear-attention (SSM) blocks expose no matching module names, and the MTP head at
layer 64 is outside the adapter — so neither was touched. That coverage is inherited from the adapter, not a choice
made here.

# 1. adapter -> GGUF
python convert_lora_to_gguf.py \
  --base-model-id Qwen/Qwen3.8-27B \
  --outfile hermes-lora-f16.gguf --outtype f16 ./lora

# 2. fuse onto an 8-bit copy of the uncensored base (higher-precision starting point than Q4)
llama-export-lora \
  -m orcarouter_Qwen3.8-27B-Uncensored-Q8_0.gguf \
  --lora hermes-lora-f16.gguf \
  -o merged.gguf

# 3. quantize
llama-quantize --allow-requantize merged.gguf \
  Qwen3.8-27B-Uncensored-hermes-Q4_K_M.gguf Q4_K_M

Q8_0 was deliberately used as the fusion base instead of the 4-bit file: the LoRA delta is small compared to the
weights (≈0.15–0.25% RMS), so fusing onto already-4-bit weights would have buried a good part of it under
quantization noise.

Usage

llama.cpp:

llama-server \
  -m Qwen3.8-27B-Uncensored-hermes-Q4_K_M.gguf \
  --mmproj mmproj-Qwen3.8-27B-Uncensored-hermes-F16.gguf \
  --ctx-size 131072 --flash-attn on \
  --cache-type-k q8_0 --cache-type-v q8_0 \
  --reasoning off

--reasoning off matters. The upstream card is explicit that this adapter is trained for thinking-off use: thinking-on
is within noise of thinking-off for the adapter and costs roughly 3× the tokens. On this base, leaving thinking on is
the single largest latency cost per reply.

LM Studio: load the Q4_K_M file, attach the mmproj file for image input, disable thinking in the model settings.

What to expect, and what not to

  • Uncensored behaviour comes from the abliterated base weights, which are present here in full. Nothing in this
    merge restores refusals.
  • The adapter's published numbers (SWE-bench Pro 43.1%, Terminal-Bench 2.1 58.8% thinking-off) were measured on the
    stock, non-abliterated base at BF16 / 131k context.
    Abliteration edits the same residual-stream components the
    adapter was trained against, so expect a partial, not exact, transfer of those gains. This specific combination has
    not been benchmarked.
  • Q4_K_M dilutes the adapter somewhat. A rank-32 delta at ~0.2% of weight RMS sits close to 4-bit quantization
    noise. Behaviour tracks the base + adapter, but a Q6_K or Q8_0 build of the same merge would be measurably closer to
    the intended model if you have the memory.
  • Not evaluated by the uploader beyond smoke tests (loads, answers, no reasoning tokens leaked at --reasoning off,
    coherent coding answers, merge deltas verified numerically against B·A·α/r).

Context

The base supports long context; 131k is a sane default for 48 GB unified memory with q8_0 KV cache. 262k works but
prefill cost and memory pressure rise sharply, and long contexts dilute attention.

Credits

Apache-2.0, inherited. Abliterated models carry no safety tuning — you are responsible for what you run them on.

README history 1 version

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

  1. 2026-11-21 Upload README.md with huggingface_hub b8cc8ed 4.9 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