mudler/holo3-35b-a3b-apex-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "apache-2.0",
    "base_model": "Hcompany/Holo3-35B-A3B",
    "tags": [
      "gguf",
      "quantized",
      "apex",
      "moe",
      "mixture-of-experts",
      "holo3",
      "vlm",
      "vision"
    ],
    "frontmatter": {
      "license": "apache-2.0",
      "base_model": "Hcompany/Holo3-35B-A3B",
      "tags": [
        "gguf",
        "quantized",
        "apex",
        "moe",
        "mixture-of-experts",
        "holo3",
        "vlm",
        "vision"
      ]
    },
    "hero_image_url": "",
    "summary": "**APEX (Adaptive Precision for EXpert Models)** quantizations of Holo3-35B-A3B. **Brought to you by the LocalAI team** | APEX Project | Technical Report",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: apache-2.0\nbase_model: Hcompany/Holo3-35B-A3B\ntags:\n  - gguf\n  - quantized\n  - apex\n  - moe\n  - mixture-of-experts\n  - holo3\n  - vlm\n  - vision\n---\n\n# Holo3-35B-A3B APEX GGUF\n\n**APEX (Adaptive Precision for EXpert Models)** quantizations of [Holo3-35B-A3B](https://huggingface.co/Hcompany/Holo3-35B-A3B).\n\n**Brought to you by the [LocalAI](https://github.com/mudler/LocalAI) team** | [APEX Project](https://github.com/mudler/apex-quant) | [Technical Report](https://github.com/mudler/apex-quant/blob/main/paper/APEX_Technical_Report.pdf)\n\n## Benchmark Results\n\nBenchmarks coming soon. For reference APEX benchmarks on the same Qwen3.5-MoE architecture, see [mudler/Qwen3.5-35B-A3B-APEX-GGUF](https://huggingface.co/mudler/Qwen3.5-35B-A3B-APEX-GGUF).\n\n## Available Files\n\n| File | Profile | Size | Best For |\n|------|---------|------|----------|\n| Holo3-35B-A3B-APEX-I-Balanced.gguf | I-Balanced | ~24 GB | Best overall quality/size ratio |\n| Holo3-35B-A3B-APEX-I-Quality.gguf | I-Quality | ~22 GB | Highest quality with imatrix |\n| Holo3-35B-A3B-APEX-Quality.gguf | Quality | 22 GB | Highest quality standard |\n| Holo3-35B-A3B-APEX-Balanced.gguf | Balanced | 24 GB | General purpose |\n| Holo3-35B-A3B-APEX-I-Compact.gguf | I-Compact | ~17 GB | Consumer GPUs, best quality/size |\n| Holo3-35B-A3B-APEX-Compact.gguf | Compact | 17 GB | Consumer GPUs |\n| Holo3-35B-A3B-APEX-I-Mini.gguf | I-Mini | ~13 GB | Smallest viable, fastest inference |\n| mmproj.gguf | Vision projector | ~858 MB | Required for image understanding |\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, technical report, and scripts.\n\n## Architecture\n\n- **Model**: Holo3-35B-A3B (Qwen3.5-MoE variant, VLM)\n- **Layers**: 40 (30 linear attention + 10 full attention)\n- **Experts**: 256 routed + 1 shared (8 active per token)\n- **Total Parameters**: ~35B\n- **Active Parameters**: ~3B per token\n- **Vision**: Built-in vision encoder (mmproj included)\n- **APEX Config**: 5+5 symmetric edge gradient across 40 layers\n\n## Run with LocalAI\n\n```bash\nlocal-ai run mudler/Holo3-35B-A3B-APEX-GGUF@Holo3-35B-A3B-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",
    "holo3",
    "vlm",
    "vision",
    "base_model:Hcompany/Holo3-35B-A3B",
    "base_model:quantized:Hcompany/Holo3-35B-A3B",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 4,
  "downloads": 3156,
  "gated": false,
  "private": false,
  "last_modified": "2026-04-04T22:50:40.000Z",
  "created_at": "2026-04-02T21:08:10.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "69cedaba0689e4caa5a9fc2f",
  "id": "mudler/Holo3-35B-A3B-APEX-GGUF",
  "modelId": "mudler/Holo3-35B-A3B-APEX-GGUF",
  "sha": "57998c466360e3c2e029c30f0c0a4ef78413cabb",
  "createdAt": "2026-04-02T21:08:10.000Z",
  "lastModified": "2026-04-04T22:50:40.000Z",
  "author": "mudler",
  "downloads": 3156,
  "likes": 4,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 10
}