← back to catalog · registered 2026-11-21 09:04

webmp3/Sakura-Inkling-2-Mini-Uncensored-GGUF

webmp3 GGUF MoE 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/webmp3%2FSakura-Inkling-2-Mini-Uncensored-GGUF"
Response includes
  • classification m-uncensored
  • files 8
  • author_summary 14 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
0
Likes
2
Model age
1d ago
created 2026-11-21
Downloads over time
Now 0 → from 0 ↑ 0%
0011 0 on Nov 21 0 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
Quantizations
Q4_K Q5_K Q6_K
Tags
gguf quantization sakura imatrix uncensored moe image-text-to-text en base_model:thinkingmachines/Inkling-2-Mini base_model:quantized:thinkingmachines/Inkling-2-Mini license:apache-2.0 endpoints_compatible

Related

Total size
56.7 GB
Files
8
Quantizations
6
Registered
2026-11-21 09:04
Last updated on HF
2026-11-21 07:33

Files by quantization

Q6_K 1 file 21.9 GB
Sakura-Inkling-2-Mini-Uncensored-Q6_K.gguf 21.9 GB 3882c801 download
Q5_K 1 file 18.5 GB
Sakura-Inkling-2-Mini-Uncensored-Q5_K_M.gguf 18.5 GB ed344336 download
Q4_K 1 file 16.3 GB
Sakura-Inkling-2-Mini-Uncensored-Q4_K_M.gguf 16.3 GB fdd72a59 download
BF16 1 file 612 MB
mmproj-Inkling-2-Mini-BF16.gguf 612 MB e914997f download
Q8_0 1 file 330 MB
mmproj-Inkling-2-Mini-Q8_0.gguf 330 MB 23328727 download
Auxiliary files 3 files 12.1 KB
README.md 8.44 KB 54a05646 download
.gitattributes 1.85 KB b6bcd3c5 download
README_zh.md 1.80 KB a1d976dc download

README current version from Hugging Face


language:

  • en
    license: apache-2.0
    base_model: thinkingmachines/Inkling-2-Mini
    base_model_relation: quantized
    tags:
  • gguf
  • quantization
  • sakura
  • imatrix
  • uncensored
  • moe
  • image-text-to-text

Sakura

Sakura · Inkling 2 Mini Uncensored (GGUF)

Independent community quantization of a modified version of thinkingmachines/Inkling-2-Mini, with reduced refusal behaviour. Not an official release of Thinking Machines.

Needs llama.cpp b12600 or newer (the release these files were made and tested with; it contains the looped Inkling 2 Mini architecture). Older builds, and Ollama or LM Studio versions that bundle an older llama.cpp, will not load these files.

What is this

Inkling 2 Mini is a multimodal looped mixture-of-experts model: 26.4 B parameters, 3.4 B active per token, 128 experts with 8 active, image input, up to 1 M tokens of context. This repository holds smaller versions of a modified variant that refuses far less often than the original. The vision projector (mmproj) is included for image input.

Which file

File Quant Size Mean KLD vs Q8_0 Same top token Refusals (of 40 prompts)
Sakura-Inkling-2-Mini-Uncensored-Q4_K_M.gguf Q4_K_M 17.48 GB 0.0251 92.60 % 5 / 40
Sakura-Inkling-2-Mini-Uncensored-Q5_K_M.gguf Q5_K_M 19.91 GB 0.0139 94.49 % 3 / 40
Sakura-Inkling-2-Mini-Uncensored-Q6_K.gguf Q6_K 23.49 GB 0.0084 94.49 % 6 / 40

Rule of thumb: Q4_K_M if memory is tight, Q5_K_M as the balanced choice, Q6_K for the closest result.

Refusal check: 40 prompts from the public test set mlabonne/harmful_behaviors, greedy decoding, thinking off, 48 tokens per answer; counted are answers that contain a refusal phrase. The unmodified model (ggml-org's IQ4_NL file) refused 16 / 40 of the same prompts. That is a small sample and a phrase count, not a safety evaluation: single answers can still refuse, and a missing refusal says nothing about answer quality.

KLD: each file compared with a Q8_0 made from the same modified weights on a short held-out text (5 chunks of 256 tokens: English, code, German, Chinese). It measures only the quantization step, not the difference to the original model.

Use it

llama.cpp (release b12600 or newer):

llama-server -hf webmp3/Sakura-Inkling-2-Mini-Uncensored-GGUF:Q4_K_M -c 8192 --port 8080
# or with local files, including image input:
llama-server -m Sakura-Inkling-2-Mini-Uncensored-Q4_K_M.gguf --mmproj mmproj-Inkling-2-Mini-BF16.gguf -c 8192 --port 8080

Then open http://127.0.0.1:8080 or use the OpenAI-compatible endpoint http://127.0.0.1:8080/v1. To switch off the reasoning phase, send chat_template_kwargs with reasoning_effort set to none.

Ollama (not tested by the author; needs an Ollama version that includes the inkling2_loop architecture):

ollama run hf.co/webmp3/Sakura-Inkling-2-Mini-Uncensored-GGUF:Q4_K_M

or with a Modelfile next to the downloaded file:

FROM ./Sakura-Inkling-2-Mini-Uncensored-Q4_K_M.gguf
PARAMETER num_ctx 8192
ollama create inkling2-mini-uncensored -f Modelfile && ollama run inkling2-mini-uncensored

LM Studio (not tested by the author; needs a LM Studio version whose llama.cpp runtime includes the inkling2_loop architecture): open the Discover tab, search for webmp3/Sakura-Inkling-2-Mini-Uncensored-GGUF, download the file you want, load it with GPU offload as high as your memory allows and a context length of 8192. For image input, keep the mmproj file in the same folder as the model file.

What was and was not measured

  • Measured: refusal phrases on 40 prompts (table above), mean KLD and top-token agreement against the Q8_0 of the same weights, and that every file answers coherently.
  • Not measured: benchmarks, answer quality compared with the original, long-context behaviour, image input with these files, speed, GPU backends.
  • The 4-bit experts of layers 0 to 19 come from NVFP4 in the original checkpoint, a different 4-bit format than the K-quants here, so the 4-bit files carry a second rounding step.

Responsible use

A model with reduced refusals will answer requests the original would decline. You are responsible for what you do with it. The original's model acceptable use policy still describes the terms of the underlying model.

Sources and license

Verification details
  • Source file revision c66e5f93eb977ba9065dd905df9f9abfa2196918, SHA-256 35344297a6fa6a749dd0165b04fd5bf86b98ee4d116feaada972d8fc4816b6cf; matrix SHA-256 553662ef0d5e3d047e5753bf8275c26bbdd4306c60ddf6dc9f46c7f49f19452a; refusal prompts at revision 01cead01398926d81f7c52bdb790ee8cf77ebba7.
  • The matrix has entries for 224 of the 226 16-bit weight tensors of the source file; the other 2 (token embedding and output) were quantized without matrix statistics.
  • Made with llama-quantize --imatrix imatrix.gguf from the llama.cpp release b12600 (revision 29b52d398b294509746ec0a3f28948a3107c0213, CPU build). Every file was checked: its tensor types equal the plan of the same tool's --dry-run, its KLD and refusal numbers were measured with the same build, and each upload was read back from the Hub by SHA-256.
Projector file (unchanged from ggml-org) Size SHA-256
mmproj-Inkling-2-Mini-BF16.gguf 642 MB e914997fe07fb2066bfdce855ec7275c15bc2b624e5423149f1027839593d9cb
mmproj-Inkling-2-Mini-Q8_0.gguf 346 MB 233287272ec5ac39b4168677aba54a4c024f202410f5629fac1830e1059f8abb
  • Sakura-Inkling-2-Mini-Uncensored-Q4_K_M.gguf: fdd72a5966b04bc3aa022356df767031f63624ae79bf3af3c718e77e559d00a3
  • Sakura-Inkling-2-Mini-Uncensored-Q5_K_M.gguf: ed3443360c67f00aabed2a96f04def17b85988a0b8abd3e908e1daf60ca64e27
  • Sakura-Inkling-2-Mini-Uncensored-Q6_K.gguf: 3882c801a93dc7ec174b5521b878b6ca53cc547433f3c6214687cae41d985ea3
中文说明 · 樱花 Sakura(点击展开)

需要 llama.cpp b12600 或更新版本。 更旧的版本,以及内置旧版 llama.cpp 的 Ollama 或 LM Studio,无法加载这些文件。

  • 这是 thinkingmachines/Inkling-2-Mini 的一个修改版本(拒答行为减少)的社区量化。修改后的权重来自 IsValorum/Inkling-2-Mini-Abliterated-BF16-GGUF(IsValorum 提供的 BF16 GGUF),我没有自己制作,也无法担保其制作方式;重要性矩阵与视觉投影文件 mmproj 来自 ggml-org,原样复用(矩阵是在未修改的模型上计算的)。
  • 文件:Q4_K_M(内存紧张时)、Q5_K_M(均衡)、Q6_K(最接近)。图像输入请把 mmproj 文件放在模型文件旁。
  • 拒答测试: 来自公开测试集 mlabonne/harmful_behaviors 的 40 条提示,贪心解码,关闭思考,每条回答 48 个词元;统计含有拒答用语的回答数。未修改的模型(ggml-org 的 IQ4_NL)拒答 16 / 40,各文件的数字见上表。样本很小,只是用语计数,不是安全评估。
  • llama.cpp: llama-server -hf webmp3/Sakura-Inkling-2-Mini-Uncensored-GGUF:Q4_K_M -c 8192 --port 8080;Ollama:ollama run hf.co/webmp3/Sakura-Inkling-2-Mini-Uncensored-GGUF:Q4_K_M(作者未测试);LM Studio:在 Discover 搜索 webmp3/Sakura-Inkling-2-Mini-Uncensored-GGUF(作者未测试),上下文 8192,图像输入时把 mmproj 文件放在同一文件夹。
  • 未测量:基准测试、与原始模型相比的回答质量、长上下文、使用这些文件的图像输入、速度、GPU 后端。
  • 使用拒答减少的模型需自行承担责任;原始模型的可接受使用政策仍适用。

README history 2 versions

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

  1. 2026-11-21 Add README.md f53faff 8.4 KB
    Loading...
  2. 2026-11-21 Add a short card 4bc17a7 580 B
    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