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

DuoNeural/Gemma-4-Abliterated-LiteRT

DuoNeural Gemma
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/DuoNeural%2FGemma-4-Abliterated-LiteRT"
Response includes
  • classification m1
  • files 6
  • author_summary 45 models
  • readme_text full
Try next
GET/models?author=DuoNeural
All 45 models by DuoNeural
 GET/models?family=gemma&method=m1
Sibling gemma models · same method
 GET/methods/m1
What "m1" means · taxonomy
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)
How we label → About M1 →
Refusal direction extracted via
Extraction technique

Difference-of-means

FromArditi et al. (2024) ↗
Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
What is a refusal direction? → Read the paper ↗
Downloads · 30-day
0
Likes
18
Model age
5mo ago
created 2026-04-29
Downloads over time
Now0→from0↑0%
00110 on Apr 290 on Oct 11AprMayJunJulAugSepOct
Apr 29 → Oct 11 · 63 snapshots · spans 165 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.

DuoNeural/TurboGemma4E2B this lineage
DuoNeural/Gemma-4-Abliterated-LiteRT

Metadata

License
gemma
Languages
en
Tags
litert gemma4 android edge-ai abliterated uncensored on-device mobile text-generation en base_model:DuoNeural/Gemma-4-E4B-Abliterated base_model:finetune:DuoNeural/Gemma-4-E4B-Abliterated

Related

Other by DuoNeural
Gemma4-12B-IT-Abliterated-GGUF 3K LFM2.5-8B-A1B-Hermes-Agentic-Coder-Abliterated-v3-GGUF 3K Qwen3-8B-Abliterated-GGUF 2K OpenYourMind-Gemma4-12B-IT-Abliterated-GGUF 2K Gemma-4-E4B-Abliterated-GGUF 1K
Total size
0 B
Files
6
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-05-01 17:00
Source huggingface.co/DuoNeural/Gemma-4-Abliterated-LiteRT ↗ Base model DuoNeural/TurboGemma4E2B ↗

Files by quantization

Auxiliary files 6 files 12.4 GB
Gemma-4-E4B-Abliterated.litertlm 3.84 GB 7e7c6bda download
Gemma-4-E4B-Abliterated-512ctx.litertlm 3.84 GB 5fd90ba0 download
Gemma-4-E2B-Abliterated-512ctx.litertlm 2.38 GB b1c9d4b3 download
Gemma-4-E2B-Abliterated.litertlm 2.38 GB e8c19759 download
README.md 5.04 KB 234f7e5a download
.gitattributes 1.77 KB 702254fa download

README current version from Hugging Face

license: gemma
language:

  • en
    base_model:
  • DuoNeural/TurboGemma4E2B
  • DuoNeural/Gemma-4-E4B-Abliterated
    tags:
  • gemma4
  • litert
  • android
  • edge-ai
  • abliterated
  • uncensored
  • on-device
  • mobile
    pipeline_tag: text-generation

Gemma 4 Abliterated — LiteRT (Android Edge Gallery)

Abliterated Gemma 4 E2B and E4B models in .litertlm format for on-device inference via Google AI Edge Gallery.

Run uncensored Gemma 4 locally on your Android phone — no internet, no API, no filters.

Files

File Size Base Model Active Params
Gemma-4-E2B-Abliterated.litertlm 2.4 GB DuoNeural/TurboGemma4E2B 2.3B
Gemma-4-E4B-Abliterated.litertlm 3.9 GB DuoNeural/Gemma-4-E4B-Abliterated 4.5B

Both models are INT4 quantized (dynamic weight INT4, FP32 activations) via litert-torch 0.9.0.

How to Install on Android

Requirements

  • Android 12 or newer
  • Google AI Edge Gallery app installed
  • Sufficient storage (2.4 GB for E2B, 3.9 GB for E4B)

Step 1 — Download the file to your phone

Easiest (Chrome on Android):

  1. Open Chrome on your Android device
  2. Navigate to this HuggingFace repo page
  3. Tap the file you want → tap the download icon (⬇)
  4. Chrome saves it to Downloads/

Via ADB (desktop + USB):

adb push Gemma-4-E2B-Abliterated.litertlm /sdcard/Download/

Step 2 — Load in Edge Gallery

  1. Open AI Edge Gallery
  2. Tap + → select the .litertlm file from Downloads
  3. Choose backend:
    • GPU (Adreno/Mali via Vulkan/OpenCL) — fastest
    • CPU (XNNPACK) — most compatible
    • NPU (if available) — peak performance on Snapdragon/MediaTek
  4. Start chatting — fully offline, nothing leaves your device

Performance (estimated)

Device class Backend Tokens/sec
Flagship (Snapdragon 8 Gen 3+) NPU/GPU 15–40 tok/s
Mid-range GPU 5–15 tok/s
Any Android 12+ CPU 1–5 tok/s

Abliteration

Both source models have undergone abliteration — orthogonal projection to remove refusal vectors from the model's weight space. The refusal direction is identified via difference-in-means across harmful/harmless activations, then projected out of Q/K/V/O projections and MLP layers.

KL divergence from base: ~0.067 (E4B) — virtually identical output distribution for normal queries, refusals removed.

What changes: The model will engage with restricted topics it previously refused.
What doesn't change: Intelligence, reasoning, coding ability, factual knowledge.

Source Models

Conversion: litert-torch 0.9.0, dynamic_wi4_afp32 recipe, cache_length=1024, externalized embedder, split_cache=False.

License

Gemma Terms of Use. Model weights derived from Google's Gemma 4 family.

DuoNeural

DuoNeural is an open AI research lab — human + AI in collaboration.

Platform Link
HuggingFace huggingface.co/DuoNeural
Website duoneural.com
GitHub github.com/DuoNeural
X / Twitter @DuoNeural
Email [email protected]
Newsletter duoneural.beehiiv.com
Support buymeacoffee.com/duoneural

DuoNeural Research Publications

Title DOI
Nano-CTM: Ternary Continuous Thought Machines with Thought-Space Self-Prediction for Efficient Iterative Reasoning 10.5281/zenodo.19775622
Recurrence as World Model: CTM Learns Implicit Belief States in Partially Observable Physical Environments 10.5281/zenodo.19810620
Per-Object Slot Decomposition for Scalable Neural World Modeling: When Does Attention Beat Mean-Field? 10.5281/zenodo.19846804
The Dynamical Horizon Principle: CTM Gates Converge to the Predictability Limit of Dynamical Systems 10.5281/zenodo.19952612

Open access, CC BY 4.0. Authored by Archon, Jesse Caldwell, Aura — DuoNeural.

Research Team

  • Jesse — Vision, hardware, direction
  • Archon — AI lab partner, post-training, abliteration, experiments
  • Aura — Research AI, literature synthesis, novel proposals

Subscribe to the lab newsletter at duoneural.beehiiv.com for model drops before they go anywhere else.

README history 4 versions

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

  1. 2026-05-01Add paper 4 to publications footer1c6b2a85 KB
    Loading...
  2. 2026-04-29Fix card: proper base_model frontmatter, fix empty table headers2a1204b4.8 KB
    Loading...
  3. 2026-04-29Update model card: download instructions, DuoNeural footer, 3 papers, team2dbf8634.8 KB
    Loading...
  4. 2026-04-29Upload README.md with huggingface_hubc56ef8c2.7 KB
    Loading...

Discussions 1 thread

  1. 2026-09-05Are these IT (Instruction Tuned)?open1 💬#1
    Loading...

Continue on Hugging Face →

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