bigatuna/qwen3.5-27b-sushi-coder-rl-gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "apache-2.0",
    "language": [
      "en"
    ],
    "pipeline_tag": "text-generation",
    "base_model": "unsloth/Qwen3.5-27B",
    "datasets": [
      "open-r1/codeforces-cots"
    ],
    "tags": [
      "gguf",
      "llama.cpp",
      "qwen3_5",
      "code",
      "rl",
      "atropos",
      "unsloth",
      "codex",
      "hermes-agent"
    ],
    "frontmatter": {
      "license": "apache-2.0",
      "language": [
        "en"
      ],
      "pipeline_tag": "text-generation",
      "base_model": "unsloth/Qwen3.5-27B",
      "datasets": [
        "open-r1/codeforces-cots"
      ],
      "tags": [
        "gguf",
        "llama.cpp",
        "qwen3_5",
        "code",
        "rl",
        "atropos",
        "unsloth",
        "codex",
        "hermes-agent"
      ]
    },
    "hero_image_url": "./bigatunaQwen3.5-27b-Sushi-Coder-RL-GGUF.png",
    "summary": "!Qwen3.5-27b-Sushi-Coder-RL-GGUF This is the 27B GGUF release in the Sushi Coder RL line. It is the bigger sibling to bigatuna/Qwen3.5-9b-Sushi-Coder-RL-GGUF, built with the same overall recipe but at 27B scale.",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: apache-2.0\nlanguage:\n- en\npipeline_tag: text-generation\nbase_model: unsloth/Qwen3.5-27B\ndatasets:\n- open-r1/codeforces-cots\ntags:\n- gguf\n- llama.cpp\n- qwen3_5\n- code\n- rl\n- atropos\n- unsloth\n- codex\n- hermes-agent\n---\n\n# Qwen3.5-27b-Sushi-Coder-RL-GGUF\n\n![Qwen3.5-27b-Sushi-Coder-RL-GGUF](./bigatunaQwen3.5-27b-Sushi-Coder-RL-GGUF.png)\n\nThis is the 27B GGUF release in the Sushi Coder RL line.\n\nIt is the bigger sibling to [`bigatuna/Qwen3.5-9b-Sushi-Coder-RL-GGUF`](https://huggingface.co/bigatuna/Qwen3.5-9b-Sushi-Coder-RL-GGUF), built with the same overall recipe but at 27B scale.\n\n## Training\n\nTraining followed this sequence:\n\n1. Start from the Unsloth 16-bit Qwen 3.5 27B base: [`unsloth/Qwen3.5-27B`](https://huggingface.co/unsloth/Qwen3.5-27B)\n2. SFT on [`open-r1/codeforces-cots`](https://huggingface.co/datasets/open-r1/codeforces-cots) using [Unsloth](https://github.com/unslothai/unsloth)\n3. Run RL on top with a custom coding environment built using [NousResearch/atropos](https://github.com/NousResearch/atropos), [OpenAI Codex](https://github.com/openai/codex), and [NousResearch/hermes-agent](https://github.com/NousResearch/hermes-agent)\n\n## Files\n\n- `qwen35-codeforces-27b-rl-step25-Q4_K_M.gguf`\n- `qwen35-codeforces-27b-rl-step25-Q8_0.gguf`\n\n## Quick Start\n\nDownload the Q4 quant:\n\n```bash\nhf download bigatuna/Qwen3.5-27b-Sushi-Coder-RL-GGUF \\\n  qwen35-codeforces-27b-rl-step25-Q4_K_M.gguf\n```\n\nDownload the Q8 quant:\n\n```bash\nhf download bigatuna/Qwen3.5-27b-Sushi-Coder-RL-GGUF \\\n  qwen35-codeforces-27b-rl-step25-Q8_0.gguf\n```\n\n## Evaluation Status\n\nThere are no 27B evals posted yet.\n\nEvals are expensive and take time, and the current plan is to post them in the next few weeks.\n\nIf you want a reference point in the meantime, you can look at the evals on the 9B sibling repo:\n\n- [`bigatuna/Qwen3.5-9b-Sushi-Coder-RL-GGUF`](https://huggingface.co/bigatuna/Qwen3.5-9b-Sushi-Coder-RL-GGUF)\n\nOpen to sponsorship if you want to help fund evals and future releases.\n\n## Notes\n\nThis repo currently contains GGUF exports only.\n",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "llama.cpp",
    "qwen3_5",
    "code",
    "rl",
    "atropos",
    "unsloth",
    "codex",
    "hermes-agent",
    "text-generation",
    "en",
    "dataset:open-r1/codeforces-cots",
    "base_model:unsloth/Qwen3.5-27B",
    "base_model:quantized:unsloth/Qwen3.5-27B",
    "license:apache-2.0",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 17,
  "downloads": 4562,
  "gated": false,
  "private": false,
  "last_modified": "2026-04-03T16:41:59.000Z",
  "created_at": "2026-04-02T22:06:45.000Z",
  "pipeline_tag": "text-generation",
  "library_name": ""
}

{
  "_id": "69cee875711ef77392be97e2",
  "id": "bigatuna/Qwen3.5-27b-Sushi-Coder-RL-GGUF",
  "modelId": "bigatuna/Qwen3.5-27b-Sushi-Coder-RL-GGUF",
  "sha": "2fc5bcd172dc466847adf0a1b0a35440965eb749",
  "createdAt": "2026-04-02T22:06:45.000Z",
  "lastModified": "2026-04-03T16:41:59.000Z",
  "author": "bigatuna",
  "downloads": 4562,
  "likes": 17,
  "gated": false,
  "private": false,
  "pipeline_tag": "text-generation",
  "library_name": "",
  "siblings_count": 5
}