WithinUsAI/Darwin-Astral-Coder-4B

1---2license: other3library_name: transformers4base_model:5  - LucidityAI/Astral-4B-Coder6  - openfree/Darwin-Qwen3-4B7  - Qwen/Qwen3-4B8tags:9  - qwen310  - mergekit11  - merge12  - text-generation-inference13  - code14  - coder15  - withinusai16language:17  - en18datasets:19  - LucidityAI/Astral-Post-Training-Dataset20pipeline_tag: text-generation21---22 23# Darwin-Astral-4B-Coder24 25**Darwin-Astral-4B-Coder** is a merged 4B-class coding model release from **WithIn Us AI**, designed for code generation, instruction-following, and practical developer-assistant workflows.26 27This repository is distributed as a standard **Transformers** checkpoint in **Safetensors** format and is positioned as a merge-based model that blends Darwin-style and Astral-style coding traits within a Qwen3-family 4B backbone.28 29## Model Summary30 31This model is intended for:32 33- code generation34- code explanation35- debugging assistance36- implementation planning37- instruction-following38- developer assistant workflows39- local or hosted coding inference40 41As a 4B-class model, it aims to balance stronger coding capability than very small models with a lighter deployment footprint than larger coder checkpoints.42 43## Base Model Lineage44 45The current repository metadata lists the following upstream model references:46 47- `LucidityAI/Astral-4B-Coder`48- `openfree/Darwin-Qwen3-4B`49- `Qwen/Qwen3-4B`50 51The visible merge configuration in the README also shows:52 53- `Qwen/Qwen3-4B-Instruct-2507` as the base model in the YAML block54- `Lucidity-AI-Astral-4B-Coder` as a merge source55- `openfree-Darwin-Qwen3-4B` as a merge source56 57These names are preserved here as shown on the repository page.58 59## Merge Details60 61According to the current README:62 63- this model is a merge of pre-trained language models64- it was created using **mergekit**65- the **SLERP** merge method was used66 67The repository also includes a visible `mergekit_config.yml`, which supports the merge-based packaging of the release.68 69## Dataset Lineage70 71The repository page currently shows the following dataset association:72 73- `LucidityAI/Astral-Post-Training-Dataset`74 75This suggests coding or post-training lineage connected to the Astral family used in the merge.76 77## Intended Use78 79Recommended use cases include:80 81- coding assistant experiments82- generating utility functions and scripts83- explaining code and technical concepts84- debugging support85- step-by-step implementation planning86- local developer tools87- hosted text-generation workflows for software tasks88 89## Suggested Use Cases90 91This model can be useful for:92 93- drafting Python, JavaScript, or general-purpose code94- proposing refactors95- generating boilerplate96- answering developer questions97- comparing implementation approaches98- producing structured technical responses99 100## Out-of-Scope Use101 102This model should not be relied on for:103 104- legal advice105- medical advice106- financial advice107- safety-critical automation108- autonomous production engineering without review109- security-critical code without expert validation110 111All generated code should be reviewed, tested, and validated before real-world deployment.112 113## Repository Contents114 115The repository currently includes standard Hugging Face model assets such as:116 117- `README.md`118- `.gitattributes`119- `added_tokens.json`120- `config.json`121- `mergekit_config.yml`122- `merges.txt`123- `model-00001-of-00002.safetensors`124- `model-00002-of-00002.safetensors`125- `model.safetensors.index.json`126- `special_tokens_map.json`127- `tokenizer.json`128- `tokenizer_config.json`129 130## Prompting Guidance131 132This model will usually work best with prompts that are:133 134- direct135- scoped to a clear task136- explicit about the language or framework137- clear about whether code, explanation, or both are wanted138- structured when step-by-step reasoning is useful139 140### Example prompt styles141 142**Code generation**143> Write a Python function that loads a JSON file, validates required keys, and returns cleaned records.144 145**Debugging**146> Explain why this code raises a KeyError and provide a safer corrected version.147 148**Implementation planning**149> Create a step-by-step plan for building a FastAPI service with authentication, logging, and tests.150 151**Refactoring**152> Refactor this function for readability and add basic error handling.153 154## Strengths155 156This model may be especially useful for:157 158- blended coding workflows159- practical developer assistance160- moderate-size local inference161- structured software-task prompting162- merge-model experimentation163- compact coder deployments164 165## Limitations166 167Like other merged 4B-class language models, this model may:168 169- hallucinate APIs or implementation details170- generate incomplete or incorrect code171- produce insecure patterns172- make reasoning mistakes on harder prompts173- require prompt iteration for best results174- need human validation before real-world use175 176## Attribution177 178**WithIn Us AI** is the publisher of this merged model release.179 180Credit for upstream assets remains with their original creators. The repository metadata and README specifically reference:181 182- `LucidityAI/Astral-4B-Coder`183- `openfree/Darwin-Qwen3-4B`184- `Qwen/Qwen3-4B`185- `Qwen/Qwen3-4B-Instruct-2507`186 187and the dataset:188 189- `LucidityAI/Astral-Post-Training-Dataset`190 191## License192 193This draft uses:194 195- `license: other`196 197If you maintain this repo, replace this with the exact license terms you want displayed and make sure they align with any upstream obligations from the referenced source models and datasets.198 199## Acknowledgments200 201Thanks to:202 203- **WithIn Us AI**204- **LucidityAI**205- **openfree**206- **Qwen**207- the **mergekit** ecosystem208- the Hugging Face platform209- the broader open-source LLM community210 211## Disclaimer212 213This model may produce inaccurate, insecure, biased, incomplete, or misleading outputs. All important generations, especially code and technical guidance, should be reviewed and tested before real-world use.