withinusai/within-us-coder-4b.gguf overview

{
  "metadata": {},
  "card_data": {
    "license": "other",
    "base_model": [
      "Qwen/Qwen3.5-4B"
    ],
    "library_name": "llama.cpp",
    "tags": [
      "gguf",
      "qwen",
      "qwen3.5",
      "code",
      "coder",
      "conversational",
      "text-generation",
      "withinusai"
    ],
    "language": [
      "en"
    ],
    "datasets": [
      "WithinUsAI/Python_GOD_Coder_50k",
      "reedmayhew/gemini-3.1-pro-2048-reasoning-1100x",
      "m-a-p/Code-Feedback",
      "crownelius/Opus-4.6-Reasoning-2100x-formatted",
      "crownelius/Opus4.6-No-Reasoning-260x",
      "crownelius/Creative_Writing_Multiturn_Enhanced",
      "HuggingFaceH4/llava-instruct-mix-vsft",
      "Roman1111111/gemini-3-pro-10000x-hard-high-reasoning"
    ],
    "model_type": "gguf",
    "inference": false,
    "frontmatter": {
      "license": "other",
      "base_model": [
        "Qwen/Qwen3.5-4B"
      ],
      "library_name": "llama.cpp",
      "tags": [
        "gguf",
        "qwen",
        "qwen3.5",
        "code",
        "coder",
        "conversational",
        "text-generation",
        "withinusai"
      ],
      "language": [
        "en"
      ],
      "datasets": [
        "WithinUsAI/Python_GOD_Coder_50k",
        "reedmayhew/gemini-3.1-pro-2048-reasoning-1100x",
        "m-a-p/Code-Feedback",
        "crownelius/Opus-4.6-Reasoning-2100x-formatted",
        "crownelius/Opus4.6-No-Reasoning-260x",
        "crownelius/Creative_Writing_Multiturn_Enhanced",
        "HuggingFaceH4/llava-instruct-mix-vsft",
        "Roman1111111/gemini-3-pro-10000x-hard-high-reasoning"
      ],
      "model_type": "gguf",
      "inference": "false"
    },
    "hero_image_url": "",
    "summary": "**WithIn-Us-Coder-4B.gguf** is a GGUF release from **WithIn Us AI**, built for local inference and coding-focused assistant use cases. It is based on **Qwen/Qwen3.5-4B** and distributed in quantized GGUF formats for efficient deployment in llama.cpp-compatible runtimes.",
    "quick_links": [],
    "benchmark_table_html": "",
    "readme_markdown": "---\nlicense: other\nbase_model:\n  - Qwen/Qwen3.5-4B\nlibrary_name: llama.cpp\ntags:\n  - gguf\n  - qwen\n  - qwen3.5\n  - code\n  - coder\n  - conversational\n  - text-generation\n  - withinusai\nlanguage:\n  - en\ndatasets:\n  - WithinUsAI/Python_GOD_Coder_50k\n  - reedmayhew/gemini-3.1-pro-2048-reasoning-1100x\n  - m-a-p/Code-Feedback\n  - crownelius/Opus-4.6-Reasoning-2100x-formatted\n  - crownelius/Opus4.6-No-Reasoning-260x\n  - crownelius/Creative_Writing_Multiturn_Enhanced\n  - HuggingFaceH4/llava-instruct-mix-vsft\n  - Roman1111111/gemini-3-pro-10000x-hard-high-reasoning\nmodel_type: gguf\ninference: false\n---\n\n# WithIn-Us-Coder-4B.gguf\n\n**WithIn-Us-Coder-4B.gguf** is a GGUF release from **WithIn Us AI**, built for local inference and coding-focused assistant use cases. It is based on **Qwen/Qwen3.5-4B** and distributed in quantized GGUF formats for efficient deployment in llama.cpp-compatible runtimes.\n\n## Model Summary\n\nThis model is intended as a coding-oriented conversational assistant with emphasis on:\n\n- code generation\n- code reasoning\n- implementation planning\n- debugging assistance\n- instruction following\n- general assistant-style chat for development workflows\n\nThis repository currently provides the following GGUF variants:\n\n- `WithIn-Us-Coder-4B.Q4_K_M.gguf`\n- `WithIn-Us-Coder-4B.Q5_K_M.gguf`\n\n## Creator\n\n**WithIn Us AI** is the creator of this model release, including the model packaging, fine-tuning / merging concept, process, naming, and GGUF distribution.\n\n## Base Model\n\nThis model is based on:\n\n- **Qwen/Qwen3.5-4B**\n\nCredit and appreciation go to the original creators of the base LLM architecture and weights.\n\n## Training Data\n\nThe current repository metadata lists the following datasets as part of the model’s training / fine-tuning lineage:\n\n- `WithinUsAI/Python_GOD_Coder_50k`\n- `reedmayhew/gemini-3.1-pro-2048-reasoning-1100x`\n- `m-a-p/Code-Feedback`\n- `crownelius/Opus-4.6-Reasoning-2100x-formatted`\n- `crownelius/Opus4.6-No-Reasoning-260x`\n- `crownelius/Creative_Writing_Multiturn_Enhanced`\n- `HuggingFaceH4/llava-instruct-mix-vsft`\n- `Roman1111111/gemini-3-pro-10000x-hard-high-reasoning`\n\n**Attribution note:**  \nWithIn Us AI does not claim ownership over third-party base models or third-party datasets. Full credit, thanks, and attribution belong to the original model and dataset creators.\n\n## Intended Use\n\nThis model is intended for:\n\n- local coding assistants\n- offline development help\n- code explanation\n- bug-fixing support\n- prompt-based code generation\n- experimentation in llama.cpp and GGUF-compatible environments\n\n### Suggested Use Cases\n\n- generating Python, JavaScript, C++, and other programming language snippets\n- explaining code blocks\n- rewriting or improving functions\n- brainstorming implementation strategies\n- creating scaffolding and prototypes\n- assisting with debugging and refactoring\n\n## Out-of-Scope Use\n\nThis model is not guaranteed to be reliable for:\n\n- high-stakes legal advice\n- medical advice\n- financial decision-making\n- autonomous execution without review\n- security-critical production decisions without human verification\n\nUsers should always validate generated code before deployment.\n\n## Quantization Formats\n\nThis repository currently includes:\n\n- **Q4_K_M** for smaller memory footprint and faster local inference\n- **Q5_K_M** for improved quality while remaining efficient\n\nChoose the quant level based on your hardware budget and quality needs.\n\n## Prompting Notes\n\nAs a coding-focused conversational model, best results usually come from prompts that are:\n\n- specific\n- structured\n- explicit about language, framework, or goal\n- clear about desired output format\n\nExample prompt style:\n\n> Write a Python function that parses a CSV file, removes duplicate rows by email, and saves the cleaned result. Include error handling and comments.\n\n## Limitations\n\nLike other language models, this model may:\n\n- hallucinate APIs or library behavior\n- generate insecure or inefficient code\n- make reasoning mistakes\n- produce outdated patterns\n- require prompt iteration for best results\n\nHuman review is strongly recommended, especially for production code.\n\n## License\n\nThis repository uses a **custom WithIn Us AI license approach**.\n\n- The base model may be subject to its original upstream license and terms.\n- Third-party datasets remain the property of their respective creators / licensors.\n- WithIn Us AI claims authorship of the fine-tuning / merging concept, process, packaging, naming, and release structure for this model distribution.\n- This repository does **not** claim ownership over third-party datasets or the underlying upstream base model.\n\nYou can include a `LICENSE` file in this repository with the exact custom terms you want enforced.\n\n## Acknowledgments\n\nSpecial thanks to:\n\n- **Qwen** for the base model\n- all third-party dataset creators listed above\n- the open-source GGUF / llama.cpp ecosystem\n- the broader Hugging Face community\n\n## Files\n\nCurrent repository files include:\n\n- `WithIn-Us-Coder-4B.Q4_K_M.gguf`\n- `WithIn-Us-Coder-4B.Q5_K_M.gguf`\n\n## Disclaimer\n\nThis model may generate incorrect, biased, insecure, or incomplete outputs.  \nUse responsibly, validate important results, and review all generated code before real-world use.",
    "related_quantizations": []
  },
  "tags": [
    "llama.cpp",
    "gguf",
    "qwen",
    "qwen3.5",
    "code",
    "coder",
    "conversational",
    "text-generation",
    "withinusai",
    "en",
    "dataset:WithinUsAI/Python_GOD_Coder_50k",
    "dataset:reedmayhew/gemini-3.1-pro-2048-reasoning-1100x",
    "dataset:m-a-p/Code-Feedback",
    "dataset:crownelius/Opus-4.6-Reasoning-2100x-formatted",
    "dataset:crownelius/Opus4.6-No-Reasoning-260x",
    "dataset:crownelius/Creative_Writing_Multiturn_Enhanced",
    "dataset:HuggingFaceH4/llava-instruct-mix-vsft",
    "dataset:Roman1111111/gemini-3-pro-10000x-hard-high-reasoning",
    "base_model:Qwen/Qwen3.5-4B",
    "base_model:quantized:Qwen/Qwen3.5-4B",
    "license:other",
    "region:us"
  ],
  "likes": 4,
  "downloads": 480,
  "gated": false,
  "private": false,
  "last_modified": "2026-03-21T21:40:25.000Z",
  "created_at": "2026-03-10T04:31:50.000Z",
  "pipeline_tag": "text-generation",
  "library_name": "llama.cpp"
}

{
  "_id": "69af9eb697e57fcb3548f085",
  "id": "WithinUsAI/WithIn-Us-Coder-4B.gguf",
  "modelId": "WithinUsAI/WithIn-Us-Coder-4B.gguf",
  "sha": "29c5663792226f7ad4e43287ccd9b306d1fda257",
  "createdAt": "2026-03-10T04:31:50.000Z",
  "lastModified": "2026-03-21T21:40:25.000Z",
  "author": "WithinUsAI",
  "downloads": 480,
  "likes": 4,
  "gated": false,
  "private": false,
  "pipeline_tag": "text-generation",
  "library_name": "llama.cpp",
  "siblings_count": 4
}