mudler/gemopus-4-26b-a4b-it-preview-apex-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "apache-2.0",
    "base_model": "Jackrong/Gemopus-4-26B-A4B-it-Preview",
    "tags": [
      "gguf",
      "quantized",
      "apex",
      "moe",
      "mixture-of-experts",
      "gemma4"
    ],
    "frontmatter": {
      "license": "apache-2.0",
      "base_model": "Jackrong/Gemopus-4-26B-A4B-it-Preview",
      "tags": [
        "gguf",
        "quantized",
        "apex",
        "moe",
        "mixture-of-experts",
        "gemma4"
      ]
    },
    "hero_image_url": "",
    "summary": "**APEX (Adaptive Precision for EXpert Models)** quantizations of Jackrong/Gemopus-4-26B-A4B-it-Preview. **Brought to you by the LocalAI team** | APEX Project",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: apache-2.0\nbase_model: Jackrong/Gemopus-4-26B-A4B-it-Preview\ntags:\n  - gguf\n  - quantized\n  - apex\n  - moe\n  - mixture-of-experts\n  - gemma4\n---\n\n# Gemopus 4 26B-A4B APEX GGUF\n\n**APEX (Adaptive Precision for EXpert Models)** quantizations of [Jackrong/Gemopus-4-26B-A4B-it-Preview](https://huggingface.co/Jackrong/Gemopus-4-26B-A4B-it-Preview).\n\n**Brought to you by the [LocalAI](https://github.com/mudler/LocalAI) team** | [APEX Project](https://github.com/mudler/apex-quant)\n\n## Available Files\n\n| File | Profile | Size | Best For |\n|------|---------|------|----------|\n| gemopus-4-26B-A4B-APEX-I-Quality.gguf | I-Quality | 20 GB | Highest quality with imatrix |\n| gemopus-4-26B-A4B-APEX-Quality.gguf | Quality | 20 GB | Highest quality standard |\n| gemopus-4-26B-A4B-APEX-I-Balanced.gguf | I-Balanced | 19 GB | Best overall quality/size ratio |\n| gemopus-4-26B-A4B-APEX-Balanced.gguf | Balanced | 19 GB | General purpose |\n| gemopus-4-26B-A4B-APEX-I-Compact.gguf | I-Compact | 15 GB | Consumer GPUs, best quality/size |\n| gemopus-4-26B-A4B-APEX-Compact.gguf | Compact | 15 GB | Consumer GPUs |\n| gemopus-4-26B-A4B-APEX-I-Mini.gguf | I-Mini | 13 GB | Smallest viable, fastest inference |\n| gemopus-4-26B-A4B-F16.gguf | F16 | 48 GB | Full precision reference |\n\n## Benchmark Results (Native Evals)\n\n| Model | Size | PPL | KL mean | HellaSwag | Winogrande | MMLU | ARC | TruthfulQA | pp512 t/s | tg128 t/s |\n|-------|------|-----|---------|-----------|------------|------|-----|------------|-----------|-----------|\n| APEX-I-Quality | 19G | 1223.5 | 0.532 | 50.5 | 59.2 | 32.1 | 35.1 | 31.0 | 5632 | 145.9 |\n| APEX-Quality | 19G | 1203.1 | 0.579 | 49.0 | 58.5 | 33.7 | 36.8 | 29.3 | 5623 | 143.5 |\n| APEX-I-Balanced | 18G | 1216.4 | 0.600 | 50.0 | 57.2 | 32.6 | 33.4 | 29.9 | 6211 | 149.4 |\n| APEX-Balanced | 18G | 1117.9 | 0.702 | 47.8 | 57.2 | 33.6 | 34.1 | 31.1 | 6221 | 145.7 |\n| APEX-I-Compact | 14G | 1258.5 | 0.943 | 49.0 | 59.0 | 32.6 | 34.1 | 30.1 | 6612 | 146.7 |\n| APEX-Compact | 14G | 782.1 | 1.617 | 48.8 | 58.2 | 33.5 | 34.4 | 30.0 | 6517 | 142.2 |\n| APEX-I-Mini | 12G | 1915.3 | 1.907 | 52.0 | 58.2 | 34.4 | 33.4 | 30.8 | 5904 | 146.8 |\n| F16 (ref) | 48G | 1215.9 | - | - | - | - | - | - | 2718 | 97.9 |\n\n## What is APEX?\n\nAPEX 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).\n\nSee the [APEX project](https://github.com/mudler/apex-quant) for full details.\n\n## Architecture\n\n- **Base Model**: [Jackrong/Gemopus-4-26B-A4B-it-Preview](https://huggingface.co/Jackrong/Gemopus-4-26B-A4B-it-Preview)\n- **Architecture**: Gemma 4 26B-A4B (MoE)\n- **Layers**: 30\n- **Experts**: 128 routed (8 active per token)\n- **Total Parameters**: 26B\n- **Active Parameters**: ~4B per token\n- **APEX Config**: 5+5 symmetric edge gradient across 30 layers\n- **Calibration**: v1.2 diverse dataset\n\n## Run with LocalAI\n\n```bash\nlocal-ai run mudler/Gemopus-4-26B-A4B-it-Preview-APEX-GGUF@gemopus-4-26B-A4B-APEX-I-Balanced.gguf\n```\n\n## Credits\n\nAPEX is brought to you by the [LocalAI](https://github.com/mudler/LocalAI) team. Developed through human-driven, AI-assisted research. Built on [llama.cpp](https://github.com/ggerganov/llama.cpp).\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "quantized",
    "apex",
    "moe",
    "mixture-of-experts",
    "gemma4",
    "base_model:Jackrong/Gemopus-4-26B-A4B-it-Preview",
    "base_model:quantized:Jackrong/Gemopus-4-26B-A4B-it-Preview",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 6,
  "downloads": 5030,
  "gated": false,
  "private": false,
  "last_modified": "2026-04-09T20:16:59.000Z",
  "created_at": "2026-04-09T19:51:13.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "69d80331d9e1eb53b9c7ea94",
  "id": "mudler/Gemopus-4-26B-A4B-it-Preview-APEX-GGUF",
  "modelId": "mudler/Gemopus-4-26B-A4B-it-Preview-APEX-GGUF",
  "sha": "58d422c8f98c77a283ca7a823bf68acc5db59800",
  "createdAt": "2026-04-09T19:51:13.000Z",
  "lastModified": "2026-04-09T20:16:59.000Z",
  "author": "mudler",
  "downloads": 5030,
  "likes": 6,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 10
}