GraySoft
Download guIDE →
Model Comparison

ccssne/gemma-4-31b-it-heretic-ara-ggufvsmudler/gemma-4-26b-a4b-it-heretic-apex-gguf

Side-by-side comparison of ccssne/gemma-4-31b-it-heretic-ara-gguf and mudler/gemma-4-26b-a4b-it-heretic-apex-gguf: downloads, license, context length, tasks, and benchmarks.

ccssne/gemma-4-31b-it-heretic-ara-gguf

CCSSNE · —

# Gemma-4-31B-It-Heretic-GGUF Quantized using Unsloth on RTX 4070 Laptop . This version includes the vision adapter (mmproj) and is optimized for local inference.

mudler/gemma-4-26b-a4b-it-heretic-apex-gguf

mudler · —

**APEX (Adaptive Precision for EXpert Models)** quantizations of gemma-4-26B-A4B-it-heretic — an **abliterated (uncensored)** version of Gemma 4, created with the Heretic tool (v1.2.0) using Arbitrary-Rank Ablation (ARA) on layers 10-30 to reduce refusals while preserving capabi…

Side-by-side Specifications

ccssne/gemma-4-31b-it-heretic-ara-ggufmudler/gemma-4-26b-a4b-it-heretic-apex-gguf
AuthorCCSSNEmudler
Pipeline Task——
Librarygguf—
Downloads48,71418,849
Likes422
LicenseUnknownUnknown
Context Length——
Created2026-04-032026-04-05
Last Modified2026-04-032026-04-05
Tags
ggufunslothgemma-4quantizedbase_model:trohrbaugh/gemma-4-31b-it-heretic-arabase_model:quantized:trohrbaugh/gemma-4-31b-it-heretic-araendpoints_compatibleregion:usconversational
ggufquantizedapexmoemixture-of-expertsgemma4vlmvisionbase_model:coder3101/gemma-4-26B-A4B-it-hereticbase_model:quantized:coder3101/gemma-4-26B-A4B-it-heretic

View full details: ccssne/gemma-4-31b-it-heretic-ara-gguf · mudler/gemma-4-26b-a4b-it-heretic-apex-gguf