← back to catalog · registered 2026-11-21 15:04

TokenAI-zer/Inkling-2-Mini-Uncensored-oQ6-mtp

TokenAI-zer MoE 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/TokenAI-zer%2FInkling-2-Mini-Uncensored-oQ6-mtp"
Response includes
  • classification m1
  • files 16
  • author_summary 3 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
0
Likes
1
Model age
today
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
Tags
mlx safetensors inkling2_loop omlx oq quantized apple-silicon mtp dspark speculative-decoding vision moe

Related

Total size
16.6 GB
Files
16
Quantizations
1
Registered
2026-11-21 15:04
Last updated on HF
2026-11-21 14:42

Files by quantization

Auxiliary files 16 files 16.7 GB
model-00001-of-00004.safetensors 4.73 GB 590459cf download
model-00003-of-00004.safetensors 4.72 GB 6821cc4b download
model-00002-of-00004.safetensors 4.70 GB 4ddccd4c download
model-00004-of-00004.safetensors 1.90 GB bfec4f84 download
mtp.safetensors 606 MB 7c3b6513 download
tokenizer.json 25.0 MB 7f332727 download
model.safetensors.index.json 88.3 KB 44834022 download
config.json 23.5 KB 15c7fc9b download
tokenizer_config.json 12.9 KB abf8fb60 download
README.md 7.90 KB 5699589c download
chat_template.jinja 6.71 KB cae99e08 download
.gitattributes 1.53 KB 52373fe2 download
LICENSE-NOTICE.md 1.40 KB 78e5ca9c download
special_tokens_map.json 529 B e4fe9735 download
processor_config.json 461 B 0e6df76a download
generation_config.json 98.0 B 1b5d49d1 download

README current version from Hugging Face


base_model: OpenUncensored/Inkling-2-Mini-Uncensored
base_model_relation: quantized
model_name: Inkling-2-Mini-Uncensored-oQ6-mtp
library_name: mlx
pipeline_tag: image-text-to-text
license: apache-2.0
license_link: https://www.apache.org/licenses/LICENSE-2.0
language:

  • en
    tags:
  • mlx
  • omlx
  • oq
  • quantized
  • apple-silicon
  • mtp
  • dspark
  • speculative-decoding
  • vision
  • moe
  • inkling2
  • inkling2_loop
  • uncensored
  • abliterated

Inkling-2-Mini-Uncensored-oQ6-mtp

Unofficial oMLX oQ6 quantization of Inkling 2 Mini Uncensored by OpenUncensored, an abliterated build of Thinking Machines' Inkling 2 Mini, for Apple Silicon. It includes the vision encoder and the bundled DSpark draft (mtp.safetensors) that oMLX Lightning MTP uses for speculative decoding. This is an MLX safetensors repository, not GGUF or PyTorch.

I am not affiliated with Thinking Machines or OpenUncensored. The model and the abliteration belong to them; see License and attribution.

It needs oMLX 0.8.1 or later (Inkling 2 support arrived in 0.8.1; this build was made with 0.8.2).

This variant

Property Value
Quantization oMLX oQ6: experts of layers 20 to 23 affine 6-bit (requantized from MXFP8); NVFP4 experts of layers 0 to 19 unchanged; attention, embedding and vocabulary head affine 8-bit, group size 64
Effective bits/weight 5.224 for language model and vision, 5.352 including the DSpark draft
Weight shards 16.05 GiB (17.23 GB) across 4 safetensors files, plus mtp.safetensors 0.59 GiB (0.64 GB)
Whole model directory 16.66 GiB (17.89 GB)
Architecture Inkling2LoopForConditionalGeneration: 24 layers (layers 6 to 17 run 3 times per token), hidden size 2304, 128 routed experts with top-8
Context limit in config 1,048,576 tokens; usable context depends on available memory
DSpark draft (MTP) 3 layers, 54 tensors, BF16, copied byte for byte from the source

Thinking Machines released Inkling 2 Mini with its routed experts already quantized (NVFP4 in layers 0 to 19, MXFP8 in layers 20 to 23). oQ never re-encodes the NVFP4 experts: at every level they pass through unchanged, so no quantization error is added to them. Attention, embedding and vocabulary head are affine 8-bit at every level. The vision encoder and the draft stay BF16, and the router and loop controllers stay FP32.

Bit distribution

Storage format Logical weights Share Tensor storage Effective bits/weight
NVFP4, group 16 20.384B 77.25% 10.679 GiB 4.50
Affine 6-bit, group 64 4.077B 15.45% 3.085 GiB 6.50
Affine 8-bit, group 64 1.572B 5.96% 1.556 GiB 8.50
BF16 0.320B 1.21% 0.595 GiB 16.00
F32 0.036B 0.14% 0.133 GiB 32.00
Total 26.389B 100% 16.047 GiB 5.224
Component Logical weights Storage
Language model 26.070B 15.45 GiB (16.59 GB)
Vision 0.318B 0.59 GiB (0.64 GB)
DSpark draft (MTP) 0.318B 0.59 GiB (0.64 GB)

These values are computed from the shipped config.json, model.safetensors.index.json and the safetensors headers. Quantization is lossy, and these variants have no published post-quantization benchmark results.

Pick a variant

oQ4-mtp oQ6-mtp oQ8-mtp
Model directory 15.71 GiB (16.87 GB) 16.66 GiB (17.89 GB) 17.49 GiB (18.78 GB)
Effective bits/weight (LM + vision) 4.915 5.224 5.494
Experts of layers 20 to 23 affine 4-bit (requantized from MXFP8) affine 6-bit (requantized from MXFP8) MXFP8 (source, unchanged)
Experts of layers 0 to 19 NVFP4 (source, unchanged) NVFP4 (source, unchanged) NVFP4 (source, unchanged)
Decode speed, one short smoke test* 65.8 tok/s 62.8 tok/s 61.7 tok/s

* Measured once on an Apple M5 Max (128 GB) with oMLX 0.8.2, Lightning MTP (DSpark) on, a 41-token coding prompt and 300 to 570 generated tokens at reasoning_effort: low, about 71% of drafted tokens accepted. A single short run, not a benchmark: speed varies with context length, prompt and draft acceptance.

The three levels differ only in the routed experts of layers 20 to 23 (about 15% of the weights), which the source ships in MXFP8: oQ4 and oQ6 requantize them to affine 4 and 6 bits, oQ8 keeps them in MXFP8 exactly as released. Everything else is identical across the three, so the sizes are close. oQ8 is the closest to the source checkpoint; oQ4 is the smallest and decodes slightly faster. All three include the DSpark draft.

Other formats:

About the source model

Inkling 2 Mini is a looped multimodal mixture-of-experts model with 26.4B total and 3.4B active parameters, text and image input, and up to 1M tokens of context. OpenUncensored removed its refusals by directional ablation applied only to the 20 BF16 attention output projections (attn.wo_ud, layers 4 to 23); the experts, the vision encoder and the DSpark draft are byte for byte the original. Their card reports MMLU 78.95% → 79.47% and 0 of 32 refusals on harmful prompts in both chat and thinking mode, measured on the source checkpoint, not on these quantizations. Method and results are in the upstream model card.

Lineage: thinkingmachines/Inkling-2-Mini → OpenUncensored/Inkling-2-Mini-Uncensored → this quant

Use with oMLX

hf download TokenAI-zer/Inkling-2-Mini-Uncensored-oQ6-mtp --local-dir ~/.omlx/models/Inkling-2-Mini-Uncensored-oQ6-mtp
omlx serve --model-dir ~/.omlx/models --port 8000

The model ID is Inkling-2-Mini-Uncensored-oQ6-mtp. Turn on Lightning MTP in the model settings to use the DSpark draft. Without it the model still runs, but decoding is not accelerated.

Reasoning effort and sampling

The shipped chat_template.jinja and generation_config.json are unchanged from the source. The template takes a reasoning_effort as a name (none, minimal, low, medium, high, xhigh, max) or a number from 0.0 to 0.99; the default is 0.9. Pass it in chat_template_kwargs, or as reasoning_effort in the request. Use none for short answers without thinking. Start with temperature 1.0 and top_p 0.95.

Responsibility

As OpenUncensored states, this model follows instructions the original would refuse, including harmful ones. You are responsible for how you use it and for complying with the law where you are. Quantization can add errors on top of the source model's own behavior.

License and attribution

Apache 2.0, as for the source and the original model. Credit for the model goes to Thinking Machines and for the abliteration to OpenUncensored; see LICENSE-NOTICE.md.

Citation

@misc{inkling-2-mini-uncensored-oq6-mtp,
  title = {Inkling-2-Mini-Uncensored-oQ6-mtp: oMLX oQ6 quantization with DSpark draft},
  author = {TokenAI-zer},
  year = {2026},
  url = {https://huggingface.co/TokenAI-zer/Inkling-2-Mini-Uncensored-oQ6-mtp},
  note = {Unofficial quantization of OpenUncensored/Inkling-2-Mini-Uncensored}
}

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 Add model card 78c466d 7.9 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