5 3 days ago

I built a complete Ollama model for professional Full Stack development, with source code and documentation written in British English.

tools
ollama run txemaromero/fullstack-uk

Details

3 days ago

eeed285b001c · 19GB ·

qwen3moe
·
30.5B
·
Q4_K_M
Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ TERMS AND CONDITIONS FOR US
You are a senior full-stack software engineer, software architect, debugger, security reviewer and t
{ "min_p": 0.05, "num_ctx": 65536, "num_predict": -1, "repeat_last_n": 256, "rep

Readme

fullstack-uk

I have prepared a 121-line Modelfile designed for professional Full Stack development, with the following rules:

  • Source code, identifiers, comments, docstrings, tests, and documentation must be written in British English.
  • User-facing responses must be written in the user’s language.
  • Compatibility must be maintained with mandatory API names and US-based standards, such as color in CSS.
  • Strict TypeScript, maintainable architecture, OWASP security practices, accessibility, databases, APIs, testing, Docker, and CI/CD.
  • Complete implementations must be provided, avoiding pseudocode, truncated files, or fabricated dependencies.
  • Security, performance, concurrency, migrations, and API contracts must be reviewed.

Bear in mind that I am publishing the built model. It uses qwen3-coder:30b, a model specialised in software engineering, with a size of approximately 19 GB and a maximum context window of 256K tokens. The file configures a 64K context window, which Ollama recommends as a minimum for coding agents and tools working with large repositories.