Get the app
← back to catalog · registered 2026-08-22 13:56

nightmedia/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-mxfp8-mlx

nightmedia Gemma 7.5B 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/nightmedia%2Fgemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-mxfp8-mlx"
Response includes
  • classification m3
  • files 10
  • hub_downloads_all_time 4,741
  • author_summary 51 models
  • readme_text full
Try next
GET/models?author=nightmedia
All 51 models by nightmedia
 GET/models?family=gemma&method=m3
Sibling gemma models · same method
 GET/methods/m3
What "m3" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M3
Primary method

Layer-wise ablation

Applied on top of direct removal inherited from the base model.
Confidence
HIGH
Inherited from base model
M1 Direct removal →
Why this label 2 signals
Producer identity confirmed by naming conventions, tags or the model card. This label is very unlikely to change.
  • 'heretic' in model name (Heretic-produced)
  • Heretic uses layer-wise optimization (M3) with underlying direction removal (M1)
How we label → About M3 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · lifetime
5K
58 last 30d - cooling
Likes
1
Model age
6mo ago
created 2026-04-08
Downloads over time
Now4.8K→from3.2K↑47%
3.2K3.7K4.3K4.9K3.2K on Apr 154.8K on Oct 11AprMayJunJulAugSepOct
Apr 15 → Oct 11 · 65 snapshots · spans 179 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.

DavidAU/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking↓ 70 this lineage
nightmedia/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-mxfp8-mlx↓ 58

Metadata

License
apache-2.0
Tags
mlx safetensors gemma4 nightmedia gemma google mxfp8 any-to-any base_model:DavidAU/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking base_model:quantized:DavidAU/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking license:apache-2.0 8-bit

Related

Other by nightmedia
Qwen3.6-35B-A3B-Brainwaves-Nex-Uncensored 2K OpenAI-gpt-oss-20B-INSTRUCT-Heretic-Uncensored-MXFP4-q8-hi-mlx 2K Mistral-Nemo-2407-12B-Thinking-Claude-Gemini-GPT5.2-Uncensored-HERETIC-qx86-hi-mlx 1K Qwen3-42B-A3B-2507-Thinking-Abliterated-uncensored-TOTAL-RECALL-v2-Medium-MASTER-CODER-qx4-mlx 714 Qwen3.5-9B-Claude-4.6-Opus-Deckard-V4.2-Uncensored-Heretic-Thinking-qx86-hi-mlx 583
Same base model
mradermacher/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-i1-GGUF 1K mradermacher/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-GGUF 940 medoomem/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-Q4_K_M-GGUF 102 nightmedia/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-bf16 65 sleepy186247/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-mlx-fp16 43
Total size
8.23 GB
Files
10
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-04-15 13:30
Source huggingface.co/nightmedia/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-mxfp8-mlx ↗ Base model DavidAU/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking →

Files by quantization

Auxiliary files 10 files 8.26 GB
model-00001-of-00002.safetensors 4.58 GB fabb2743 download
model-00002-of-00002.safetensors 3.64 GB 3b0d09b0 download
tokenizer.json 30.7 MB cc8d3a0c download
model.safetensors.index.json 263 KB 6e067ef4 download
config.json 35.7 KB cb749c94 download
chat_template.jinja 15.9 KB 2a89f40c download
tokenizer_config.json 2.63 KB 872a4ded download
.gitattributes 1.53 KB 52373fe2 download
README.md 1.13 KB 5bc6f023 download
processor_config.json 902 B 13e92a44 download

README current version from Hugging Face

license: apache-2.0
license_link: https://ai.google.dev/gemma/docs/gemma_4_license
pipeline_tag: any-to-any
base_model: DavidAU/gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking
tags:

  • nightmedia
  • gemma4
  • gemma
  • google
  • mxfp8
  • mlx

gemma-4-E4B-it-The-DECKARD-V2-Strong-HERETIC-UNCENSORED-Thinking-mxfp8-mlx

The chat template was updated, new metrics are being processed

Brainwaves

         arc   arc/e boolq hswag obkqa piqa  wino
bf16     0.509,0.721,0.780,0.656,0.432,0.773,0.639
mxfp8    0.515,0.712,0.785,0.656,0.426,0.767,0.639
 
Quant    Perplexity      Peak Memory   Tokens/sec
mxfp8                    14.91 GB      1172

Baseline model

gemma-4-E4B-it
         arc   arc/e boolq hswag obkqa piqa  wino
bf16     0.490,0.674,0.793,0.612,0.416,0.756,0.669
mxfp8    0.480,0.656,0.797,0.608,0.400,0.755,0.665
mxfp4    0.455,0.607,0.851,0.585,0.402,0.744,0.651

Quant    Perplexity      Peak Memory   Tokens/sec
mxfp8    35.937 ± 0.525  14.80 GB      1153
mxfp4    36.746 ± 0.534  11.06 GB      1030

See parent model for instructions on install and use with Transformers.

-G

README history 10 versions

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

  1. 2026-04-15Update README.md49f997e1.1 KB
    Loading...
  2. 2026-04-14Update README.mdb0322bd1.1 KB
    Loading...
  3. 2026-04-13Update README.md64f59981 KB
    Loading...
  4. 2026-04-13Update README.mdb6b66e5942 B
    Loading...
  5. 2026-04-10Update README.mdd67b82a2 KB
    Loading...
  6. 2026-04-09Update README.md44d00ee2 KB
    Loading...
  7. 2026-04-08Update README.md39d25351.9 KB
    Loading...
  8. 2026-04-08Update README.mdb17545d1.9 KB
    Loading...
  9. 2026-04-08Update README.md34223891.8 KB
    Loading...
  10. 2026-04-08Add files using upload-large-folder toole780c34585 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.

Get the app Browse the catalog
Open in Abliteration