GitHub radar

Run a local instance of this free AI code generator

An offline AI code helper which doesn’t need any internet connection, subscription or token payment!

01ukisai/Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF 177231k downloads/mo27B paramstext-generation

A compressed neural network file for programming and multi-step agent tasks. You can install it on an ordinary computer and use it without an internet connection.

When it helps

Good for routine tasks, experiments, and as a backup when you don't have a paid subscription. For serious or complex work, stick with top-tier paid models.

Pros

  • Completely free, no subscription or vendor servers required
  • Answers faster and more concisely than the original Qwen model
  • Runs locally on an ordinary PC with 16GB of memory

Cons

  • Noticeably weaker than top models like Opus 5 or Fable 5.1
  • Not suited for complex or high-stakes work tasks

How to set it up — step by step

  1. 1Open your AI agent (Claude Code, Codex) inside your project folder
  2. 2Send the agent the text from the block below — it will download and set up the model
  3. 3Make sure your computer has at least 16GB of RAM
  4. 4Let the agent pick a compressed model file sized for your memory — 8-9 to 12GB on disk
  5. 5Check that the agent launched the model via llama.cpp or Ollama and can talk to it
  6. 6Try the model on a simple coding task before trusting it with anything important

Text for your agent

Copy this and send it to your agent — Claude Code, Codex, any of them:

Set up this model locally for me: https://huggingface.co/ukisai/Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF — pick the compressed file that fits my memory, run it through llama.cpp or Ollama, and show me how to talk to it.

A regular computer with 16 GB of memory — the smallest version takes about 8-9 GB on disk, the more accurate one about 12 GB.

Open on Hugging Face