← back to catalog · registered 2026-11-18 19:04

abacusai/Smaug-Uncensored-GGUF

abacusai 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/abacusai%2FSmaug-Uncensored-GGUF"
Response includes
  • classification m-uncensored
  • files 13
  • author_summary 6 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M-U
Primary method

Uncensored (method unknown)

No other method signals detected in this model.
Confidence
LOW
Why this label 3 signals
Weak or ambiguous signals. Best guess based on catalog patterns; treat as tentative and check the evidence below.
  • 'uncensored' in name/tags but no 'abliterated' marker
  • method not identifiable from author declaration alone
  • may be DPO fine-tune, prompt engineering, or unknown technique
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
2K
Likes
2
Model age
3d ago
created 2026-11-18
Downloads over time
Now 1.5K → from 0 ↑ 0%
05601.1K1.7K 0 on Nov 18 1.5K 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 · 2K downloads combined

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

Metadata

License
apache-2.0
Languages
en zh
Quantizations
IQ3 IQ4 Q3_K Q4_K Q5_K Q6_K Q8_0
Tags
gguf llama.cpp mtp agentic smaug abacusai image-text-to-text en zh base_model:abacusai/Smaug-Uncensored base_model:quantized:abacusai/Smaug-Uncensored license:apache-2.0

Related

Total size
119 GB
Files
13
Quantizations
10
Registered
2026-11-18 19:04
Last updated on HF
2026-11-18 18:32

Files by quantization

Q8_0 1 file 26.6 GB
Smaug-Uncensored-27b-Q8_0.gguf 26.6 GB 59515a7b download
Q6_K 1 file 20.6 GB
Smaug-Uncensored-27b-Q6_K.gguf 20.6 GB 7a1b6924 download
Q5_K 1 file 17.9 GB
Smaug-Uncensored-27b-Q5_K_M.gguf 17.9 GB edf52e56 download
Q4_K 1 file 15.4 GB
Smaug-Uncensored-27b-Q4_K_M.gguf 15.4 GB d0508815 download
IQ4 1 file 14.0 GB
Smaug-Uncensored-27b-IQ4_XS.gguf 14.0 GB 2bc554ce download
Q3_K 1 file 12.4 GB
Smaug-Uncensored-27b-Q3_K_M.gguf 12.4 GB 39888a53 download
IQ3 1 file 11.7 GB
Smaug-Uncensored-27b-IQ3_M.gguf 11.7 GB 6785e2f1 download
Drafts and companions 1 file 26.0 MB
imatrix.gguf 26.0 MB 61f2cd37 download
F16 1 file 885 MB
mmproj-Smaug-Uncensored-27b-F16.gguf 885 MB 01c8b8cd download
Auxiliary files 4 files 34.3 KB
README.md 19.7 KB 7d2ac950 download
LICENSE 11.3 KB f938136e download
.gitattributes 2.59 KB 14f807ae download
NOTICE 811 B b8cadab3 download

README current version from Hugging Face


license: apache-2.0
base_model: abacusai/Smaug-Uncensored
base_model_relation: quantized
pipeline_tag: image-text-to-text
language:

  • en
  • zh
    tags:
  • gguf
  • llama.cpp
  • mtp
  • agentic
  • smaug
  • abacusai

Smaug-Uncensored

Homepage Hugging Face
License

1. Introduction

GGUF builds of Smaug-Uncensored for llama.cpp, Ollama and LM Studio: seven sizes from IQ3_M to Q8_0, each as a plain file and as a file that keeps the model's multi-token-prediction (MTP) draft head for faster generation, plus the vision projector for image input. Every file was loaded and tested with thinking on, thinking off and an image, and every plain file plus the MTP Q6_K file was run on GPQA Diamond and IFBench (§4). What the model does and how it was trained is on the main card.

Built with llama.cpp 71a6a11 from the BF16 weights, with an importance matrix from bartowski's calibration_datav3 text (imatrix.gguf, included).

2. Which file to pick

QuantPlain fileWith MTP headMean KLD99% KLDSame top tokenPick it when you have
IQ3_M11.7 GiB11.9 GiB0.06110.62890.3%Smallest file. 16 GB of GPU memory with a 32K context; a 12 GB card needs part of the model on the CPU.
Q3_K_M12.4 GiB12.6 GiB0.05210.52790.7%16 GB of GPU memory with a context up to about 24K. Closer to full precision than IQ3_M.
IQ4_XS14.0 GiB14.3 GiB0.01790.19394.8%16 GB of GPU memory with a short context.
Q4_K_M15.4 GiB15.7 GiB0.01350.13895.4%The default (Ollama picks it). 24 GB of GPU memory.
Q5_K_M17.9 GiB18.2 GiB0.00550.05497.1%24 GB with room for a longer context.
Q6_K20.6 GiB20.9 GiB0.00280.02498.1%32 GB, or a 24 GB card with a short context.
Q8_026.6 GiB27.1 GiB0.00090.00998.8%Close to full precision. 32 GB and up, or a 48 GB Mac.
KL divergence of each plain file from the BF16 GGUF of the same weights, on wikitext-2 test (16 chunks of 2,048 tokens), measured with llama.cpp 71a6a11 llama-perplexity --kl-divergence; lower is closer. "Same top token" is how often the quant's most likely next token equals BF16's. The MTP files quantize the language model exactly like the plain files (the Q4_K_M pair was checked: all 851 shared tensors are byte-identical) and add the 15 tensors of the draft head. Memory needed is about the file size plus 64 KiB per token of context (16 full-attention layers) plus 0.15 GiB per parallel sequence for the linear-attention state, plus 0.9 GiB for the vision projector when images are used.

3. Quick start

llama.cpp (downloads the vision projector too):

llama-server -hf abacusai/Smaug-Uncensored-GGUF:Q4_K_M --jinja -c 32768 \
  --temp 1.0 --top-p 0.95 --top-k 20 --min-p 0

The model thinks by default (reasoning effort xhigh). To turn thinking off, add --chat-template-kwargs '{"enable_thinking": false}', or send "chat_template_kwargs": {"enable_thinking": false} with a request.

Faster generation with the MTP head (llama.cpp): use a file from MTP/ and turn on MTP drafting.

llama-server --hf-repo abacusai/Smaug-Uncensored-GGUF --hf-file MTP/Smaug-Uncensored-27b-MTP-Q4_K_M.gguf \
  --mmproj-url https://huggingface.co/abacusai/Smaug-Uncensored-GGUF/resolve/main/mmproj-Smaug-Uncensored-27b-F16.gguf \
  --jinja -c 32768 --spec-type draft-mtp --temp 1.0 --top-p 0.95 --top-k 20 --min-p 0

Ollama:

ollama run hf.co/abacusai/Smaug-Uncensored-GGUF:Q4_K_M

LM Studio: search for abacusai/Smaug-Uncensored-GGUF and pick a file from the table above.

4. Checks on the quantized files

GPQA Diamond and IFBench by file

Scored question by question against Q8_0 (paired exact tests), no file differs on GPQA Diamond (p 0.15 or higher). On IFBench every file scores below Q8_0's 86.0, by 1.0 to 5.3 points; the gap is significant for plain Q6_K (80.7, p 0.01) and Q3_K_M (81.0, p 0.04). The MTP Q6_K file carries the same quantized language-model weights as the plain Q6_K and scored 84.0, 3.3 points above it, and the smallest file, IQ3_M, scored 84.7, so part of these single-run gaps is sampling variation.

Q8_0
plain
Q6_K
plain
Q6_K
with MTP head
Q5_K_M
plain
Q4_K_M
plain
IQ4_XS
plain
Q3_K_M
plain
IQ3_M
plain
GPQA Diamond
198 questions
86.487.483.886.987.490.484.384.8
IFBench
300 prompts
86.080.784.082.083.085.081.084.7
Accuracy in percent, one run per file, thinking on (the template default, reasoning effort xhigh), temperature 1.0, top_p 0.95, top_k 20, min_p 0, seed 42, up to 150,000 output tokens, llama-server from llama.cpp 71a6a11. GPQA Diamond is zero-shot chain of thought; an answer counts when the last lettered choice or the boxed choice in the final answer is the key. IFBench is prompt-level loose accuracy, the main card's IFBench metric. The main card's BF16 scores come from vLLM, a different serving stack, so compare the files with each other rather than with those.

MTP draft head

The MTP head is inherited from the base model, and it still predicts this model's tokens well. With 3 draft tokens in vLLM 0.32.0, this model accepts 62.2% of drafted tokens (mean accepted length 2.87 tokens per step) against 57.9% (2.74) for the base model, on the same 150 prompts (IFEval, GSM8K, HumanEval) with thinking on. In llama.cpp's smoke test of the MTP Q4_K_M file, 123 of 165 drafted tokens were accepted.

Chat template

llama.cpp renders this repo's embedded chat template exactly as the Hugging Face tokenizer does (checked for a thinking-on turn, a thinking-off turn and a tool-call round trip, on both the plain and the MTP Q4_K_M file), so tool calling and the thinking switch behave as in the BF16 release.

5. License

Apache 2.0, inherited from Qwen3.8-27B. See LICENSE and NOTICE.

6. Citation

@misc{abacusai2026smaugUncensored,
  title  = {Smaug-Uncensored},
  author = {Abacus.AI},
  year   = {2026},
  note   = {Fine-tune of Qwen/Qwen3.8-27B with the base model's refusals removed},
  url    = {https://huggingface.co/abacusai/Smaug-Uncensored}
}

7. Contact Us

If you have any questions, please reach out at Abacus.AI.

README history 2 versions

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

  1. 2026-11-18 Card: repository renamed to Smaug-Uncensored 25cc0d0 19.7 KB
    Loading...
  2. 2026-11-18 Smaug-Uncensored GGUF release 7e72df4 19.7 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.

Open in Abliteration