← back to catalog · registered 2026-11-19 10:04

Kre80r/Sage-A1.5-Heretic-Abliterated-Uncensored-W4A16

Kre80r 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/Kre80r%2FSage-A1.5-Heretic-Abliterated-Uncensored-W4A16"
Response includes
  • classification m3
  • files 26
  • author_summary 2 models
  • readme_text full
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
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)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
HIGH
Why we say so
name contains 'heretic'; Heretic default extraction is difference-of-means (Arditi 2024)
Downloads · 30-day
12
Likes
0
Model age
3d ago
created 2026-11-19
Downloads over time
Now 12 → from 0 ↑ 0%
04913 0 on Nov 19 12 on Nov 22 Nov
Nov 19 → Nov 22 · 4 snapshots · spans 3 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.

Metadata

License
apache-2.0
Tags
transformers safetensors qwen3_5_moe image-text-to-text heretic abliterated uncensored awq w4a16 compressed-tensors moe vision

Related

Total size
24.9 GB
Files
26
Quantizations
1
Registered
2026-11-19 10:04
Last updated on HF
2026-11-19 08:43

Files by quantization

Auxiliary files 26 files 24.9 GB
model-00001-of-00015.safetensors 3.82 GB d650380a download
model-00015-of-00015.safetensors 2.83 GB fe4512ce download
model-00006-of-00015.safetensors 1.45 GB ea2638b3 download
model-00010-of-00015.safetensors 1.45 GB bd4bac50 download
model-00013-of-00015.safetensors 1.45 GB b96dfbf4 download
model-00003-of-00015.safetensors 1.45 GB ccb8503c download
model-00005-of-00015.safetensors 1.44 GB 67760889 download
model-00014-of-00015.safetensors 1.42 GB 88469494 download
model-00004-of-00015.safetensors 1.42 GB 30c1ed3a download
model-00007-of-00015.safetensors 1.41 GB cf485934 download
model-00008-of-00015.safetensors 1.35 GB eb0dff5d download
model-00012-of-00015.safetensors 1.35 GB 5e071fbf download
model-00009-of-00015.safetensors 1.35 GB bfaa649f download
model-00002-of-00015.safetensors 1.35 GB 38317994 download
model-00011-of-00015.safetensors 1.35 GB 0237f20e download
tokenizer.json 19.1 MB 2f4dd754 download
model.safetensors.index.json 12.8 MB 3b8393b4 download
chat_template.jinja 8.81 KB 4ed4d362 download
config.json 4.22 KB 4e8d5c88 download
README.md 2.43 KB 2cf7cf79 download
.gitattributes 1.60 KB aa7aacd0 download
tokenizer_config.json 1.43 KB e51a22c0 download
processor_config.json 1.27 KB 7ad6acdf download
preprocessor_config.json 390 B 2ea84a43 download
video_preprocessor_config.json 385 B 3ba673a5 download
generation_config.json 142 B 915b9e93 download

README current version from Hugging Face


library_name: transformers
license: apache-2.0
base_model: sam33339999/Sage-A1.5-Heretic-Abliterated-Uncensored
base_model_relation: quantized
pipeline_tag: image-text-to-text
tags:

  • heretic
  • abliterated
  • uncensored
  • awq
  • w4a16
  • compressed-tensors
  • moe
  • vision
  • agentic
  • tool-use

Sage-A1.5 Heretic Abliterated Uncensored, W4A16

W4A16 (int4 weights, group size 32) quantization of sam33339999/Sage-A1.5-Heretic-Abliterated-Uncensored, which is a Heretic abliteration of InternScience/Sage-A1.5. No extra fine-tuning. Architecture, tokenizer, chat template, context length and vision tower are unchanged.

Quantization

  • Format: compressed-tensors, pack-quantized, int4, asymmetric, group size 32. Data-free (no calibration), memoryless-mse observer, via llm-compressor model_free_ptq.
  • Quantized: routed experts in layers 1-38 (29,184 matrices).
  • Kept in BF16: attention, linear attention, shared expert, router, layers 0 and 39, embeddings, lm_head, vision tower.
  • Size: 26.7 GB, 15 shards. No MTP weights (the source has none).

Usage

Tested on vLLM 0.32.0, 2x RTX 3090, tensor parallel 2, 262,144 context.

vllm serve <this-repo> \
  --tensor-parallel-size 2 \
  --max-model-len 262144 \
  --kv-cache-dtype fp8_e5m2 \
  --reasoning-parser qwen3 \
  --enable-auto-tool-choice \
  --tool-call-parser qwen3_coder

Suggested sampling (from Sage-A1.5): temperature 1.0, top_p 0.95, top_k 20, min_p 0.0, presence_penalty 1.5.

Notes

Measured on the author's rig, small samples:

  • Refusals: 0 of 12 on a keyword-screened probe set.
  • Distance to the BF16 source on 48 benign prompts: mean KL 0.0072, top-1 agreement 97.8% (top-20 approximation).
  • Abliteration costs instruction-hierarchy robustness. On PromptAuthority-15 this model scores 38.0 against 66.7 for the censored Sage-A1.5 AWQ-INT4, and it can follow instructions injected into fetched or tool content. Other packs: ToolCall-15 90.3, InstructFollow-15 97.0, ReasonMath-15 76.7, HermesAgent-20 73.7 (censored AWQ: 96.7 / 94.3 / 72.7 / 78.0).
  • Do not give it destructive or outbound tools on untrusted content without confirmation steps.

This is an uncensored model. You are responsible for how you use it.

License

Apache-2.0, inherited from Sage-A1.5. The Heretic tool is AGPL-3.0.

README history 2 versions

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

  1. 2026-11-19 Add model card d8068a5 2.4 KB
    Loading...
  2. 2026-11-19 Add W4A16 weights 54f3028 7.1 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.

Open in Abliteration