GraySoft
Projects Models Compare Cloud benchmarks FAQ Download guIDE →
Model Intelligence Sheet

noctrex/LFM2.5-2.6B-heretic-uncensored-GGUF overview

These are quantizations of the model LiquidAI / LFM2.5 2.6B https://huggingface.co/LiquidAI/LFM2.5 2.6B . Abliterated with heretic 1.4.0 The quantizations were…

ggufhereticuncensoredabliteratedtext-generationbase_model:LiquidAI/LFM2.5-2.6Bbase_model:quantized:LiquidAI/LFM2.5-2.6Bendpoints_compatibleregion:usconversational

Runs locally from ~2.3 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
0
Likes
0
Pipeline
text-generation
Author

Repository Files & Downloads

9 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
LFM2.5-2.6B-heretic-uncensored-BF16.ggufGGUFBF165.03 GBDownload
LFM2.5-2.6B-heretic-uncensored-F16.ggufGGUFF165.03 GBDownload
LFM2.5-2.6B-heretic-uncensored-IQ4_NL.ggufGGUFIQ4_NL1.48 GBDownload
LFM2.5-2.6B-heretic-uncensored-IQ4_XS.ggufGGUFIQ4_XS1.41 GBDownload
LFM2.5-2.6B-heretic-uncensored-Q4_K_M.ggufGGUFQ4_K_M1.56 GBDownload
LFM2.5-2.6B-heretic-uncensored-Q5_K_M.ggufGGUFQ5_K_M1.81 GBDownload
LFM2.5-2.6B-heretic-uncensored-Q6_K.ggufGGUFQ6_K2.07 GBDownload
LFM2.5-2.6B-heretic-uncensored-Q8_0.ggufGGUFQ8_02.68 GBDownload
imatrix.ggufGGUFGGUF2.3 MBDownload

Model Details

Model IDnoctrex/LFM2.5-2.6B-heretic-uncensored-GGUF
Authornoctrex
Pipelinetext-generation
License
Base modelLiquidAI/LFM2.5-2.6B
Last modified2026-08-04T22:52:34.000Z

Model README

---

pipeline_tag: text-generation

base_model:

  • LiquidAI/LFM2.5-2.6B

tags:

  • heretic
  • uncensored
  • abliterated

---

These are quantizations of the model LiquidAI / LFM2.5-2.6B.

Abliterated with heretic 1.4.0

The quantizations were created using an imatrix merged from combined\_en\_medium and harmful.txt to leverage the abliterated nature of the model.

| Metric | This model | Original model |

| :----- | :--------: | :---------------------------: |

| Refusals | 7/100 | 96/100 |

| KL divergence | 0.0146 | 0 |

Run noctrex/LFM2.5-2.6B-heretic-uncensored-GGUF with guIDE

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE → · Browse 524k+ models · Compare models

Source: Hugging Face · Compare models