---

license: gemma

base_model: coder3101/gemma-4-26B-A4B-it-heretic

tags:

- gguf

- quantized

- apex

- moe

- mixture-of-experts

- gemma4

- vlm

- vision

---

<!-- apex-banner-v2 -->

<div style="background-color: #f59e0b; color: white; padding: 20px; border-radius: 10px; text-align: center; margin: 20px 0;">

<h2 style="color: white; margin: 0 0 10px 0;">⚡ Each donation = another big MoE quantized</h2>

<p style="font-size: 18px; margin: 0 0 15px 0;">I host <b>25+ free APEX MoE quantizations</b> as independent research. My only local hardware is an <b>NVIDIA DGX Spark</b> (122 GB unified memory) — enough for ~30-50B-class MoEs, but <b>bigger ones (200B+) require rented compute</b> on H100/H200/Blackwell, typically $20-100 per quant.<br>If APEX quants are useful to you, your support directly funds those bigger runs.</p>

<p style="font-size: 20px; margin: 0;">

<a href="https://www.patreon.com/cw/mudler" style="color: white; text-decoration: underline;">🎉 Patreon (Monthly)</a> &nbsp;|&nbsp;

<a href="https://www.buymeacoffee.com/mudler" style="color: white; text-decoration: underline;">☕ Buy Me a Coffee</a> &nbsp;|&nbsp;

<a href="https://github.com/sponsors/mudler" style="color: white; text-decoration: underline;">⭐ GitHub Sponsors</a>

</p>

<p style="font-size: 14px; margin: 10px 0 0 0; opacity: 0.9;">💚 Big thanks to Hugging Face for generously donating additional storage — much appreciated.</p>

</div>

Gemma 4 26B-A4B Heretic (Abliterated) APEX GGUF

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 capabilities (KL divergence 0.0499 from original).

Brought to you by the LocalAI team | APEX Project | Technical Report

Benchmark Results

Benchmarks coming soon. For reference APEX benchmarks on the Qwen3.5-35B-A3B architecture, see mudler/Qwen3.5-35B-A3B-APEX-GGUF.

Available Files

| File | Profile | Size | Best For |

|------|---------|------|----------|

| gemma-4-26B-A4B-heretic-APEX-I-Balanced.gguf | I-Balanced | ~19 GB | Best overall quality/size ratio |

| gemma-4-26B-A4B-heretic-APEX-I-Quality.gguf | I-Quality | ~20 GB | Highest quality with imatrix |

| gemma-4-26B-A4B-heretic-APEX-Quality.gguf | Quality | ~20 GB | Highest quality standard |

| gemma-4-26B-A4B-heretic-APEX-Balanced.gguf | Balanced | ~19 GB | General purpose |

| gemma-4-26B-A4B-heretic-APEX-I-Compact.gguf | I-Compact | ~15 GB | Consumer GPUs, best quality/size |

| gemma-4-26B-A4B-heretic-APEX-Compact.gguf | Compact | ~15 GB | Consumer GPUs |

| gemma-4-26B-A4B-heretic-APEX-I-Mini.gguf | I-Mini | ~13 GB | Smallest viable, fastest inference |

| mmproj.gguf | Vision projector | ~1.2 GB | Required for image understanding |

What is APEX?

APEX is a quantization strategy for Mixture-of-Experts (MoE) models. It classifies tensors by role (routed expert, shared expert, attention) and applies a layer-wise precision gradient -- edge layers get higher precision, middle layers get more aggressive compression. I-variants use diverse imatrix calibration (chat, code, reasoning, tool-calling, agentic traces, Wikipedia).

See the APEX project for full details, technical report, and scripts.

Architecture

• Model: gemma-4-26B-A4B-it-heretic (same architecture as gemma-4-26B-A4B-it)
• Layers: 30
• Experts: 128 routed (8 active per token)
• Total Parameters: 26B
• Active Parameters: ~4B per token
• Vision: Built-in vision encoder (mmproj included)
• APEX Config: 5+5 symmetric edge gradient across 30 layers
• Calibration: v1.3 diverse dataset

Run with LocalAI

local-ai run mudler/gemma-4-26B-A4B-it-heretic-APEX-GGUF@gemma-4-26B-A4B-heretic-APEX-I-Balanced.gguf

Credits

APEX is brought to you by the LocalAI team. Developed through human-driven, AI-assisted research. Built on llama.cpp.