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

tgncr/typhoon2.5-qwen3-30b-a3b-abliterated

tgncr Qwen 31B MoE
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/tgncr%2Ftyphoon2.5-qwen3-30b-a3b-abliterated"
Response includes
  • classification m1
  • files 10
  • hub_downloads_all_time 732
  • author_summary 2 models
  • readme_text full
Try next
GET/models?author=tgncr
All 2 models by tgncr
 GET/models?family=qwen&method=m1
Sibling qwen 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 · lifetime
732
15 last 30d - cooling
Likes
0
Model age
8mo ago
created 2026-02-06
Downloads over time
Now737→from325↑127%
304462620778325 on Feb 4737 on Oct 11FebAprJunAugOct
Feb 4 → Oct 11 · 75 snapshots · spans 249 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.

typhoon-ai/typhoon2.5-qwen3-30b-a3b this lineage
tgncr/typhoon2.5-qwen3-30b-a3b-abliterated↓ 15

Variants by this author 2 formats · 229 downloads combined

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

Metadata

Tags
transformers safetensors qwen3_moe text-generation heretic uncensored decensored abliterated conversational base_model:typhoon-ai/typhoon2.5-qwen3-30b-a3b base_model:finetune:typhoon-ai/typhoon2.5-qwen3-30b-a3b endpoints_compatible

Related

Other by tgncr
typhoon2.5-qwen3-30b-a3b-abliterated-GGUF 214
Total size
56.9 GB
Files
10
Quantizations
1
Registered
2026-08-22 13:56
Last updated on HF
2026-02-06 06:48
Source huggingface.co/tgncr/typhoon2.5-qwen3-30b-a3b-abliterated ↗ Base model typhoon-ai/typhoon2.5-qwen3-30b-a3b ↗

Files by quantization

Auxiliary files 10 files 56.9 GB
model-00001-of-00002.safetensors 46.4 GB 4640874e download
model-00002-of-00002.safetensors 10.4 GB 6ae8b32e download
tokenizer.json 10.9 MB be756060 download
model.safetensors.index.json 1.62 MB cb2176f6 download
chat_template.jinja 2.57 KB 70adff8a download
.gitattributes 1.53 KB 52373fe2 download
README.md 1.11 KB 0c68b805 download
config.json 1020 B bdddeecd download
tokenizer_config.json 705 B feb42f59 download
generation_config.json 120 B aa21bbb8 download

README current version from Hugging Face

library_name: transformers
tags:

  • heretic
  • uncensored
  • decensored
  • abliterated
    base_model:
  • typhoon-ai/typhoon2.5-qwen3-30b-a3b

This is a decensored version of typhoon-ai/typhoon2.5-qwen3-30b-a3b, made using Heretic v1.1.0

Abliteration parameters

Parameter Value
direction_index per layer
attn.o_proj.max_weight 1.34
attn.o_proj.max_weight_position 28.51
attn.o_proj.min_weight 0.99
attn.o_proj.min_weight_distance 16.42
mlp.down_proj.max_weight 1.33
mlp.down_proj.max_weight_position 45.00
mlp.down_proj.min_weight 0.36
mlp.down_proj.min_weight_distance 16.47

Performance

Metric This model Original model (typhoon-ai/typhoon2.5-qwen3-30b-a3b)
KL divergence 0.1273 0 (by definition)
Refusals 4/100 100/100

Enjoy mate 🧙🏻‍♂️!

README history 4 versions

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

  1. 2026-02-06Update README.md30f07cf1.1 KB
    Loading...
  2. 2026-02-06Update README.md6651ec76.1 KB
    Loading...
  3. 2026-02-06Upload README.md with huggingface_hubedd69516.1 KB
    Loading...
  4. 2026-02-06Upload Qwen3MoeForCausalLM800e9a55.1 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.

Get the app Browse the catalog
Open in Abliteration