GitHub radar
UkisAI opens free API for Swift-Qwen3.8-27B vision tasks
UkisAI built a faster-reasoning derivative of Alibaba's Qwen3.8-27B and opened a free OpenAI-compatible API — the model handles text and images with a 262K-token context.
Swift-Qwen3.8-27B is a reasoning-efficient derivative of Qwen3.8-27B from Alibaba, built by UkisAI. The base model handles text and images, with a context window of 262,144 tokens. UkisAI identified tokens that trigger excessive reasoning chains, then fine-tuned the model to use them less often. By their measurements, the result uses 58.3% fewer thinking tokens and is 1.95x faster, with less than 1% performance loss. UkisAI provides a free OpenAI-compatible API at ukisai.com for individuals and companies with annual revenue under $1 million.
Why a vibe-coder should care
For agents that need to analyze images or process documents via API, this is a free alternative to paid providers. The endpoint is OpenAI-compatible, so existing agent code needs minimal changes. For complex reasoning tasks this is not a replacement for frontier models. For simple and repetitive work — describing screenshots, classifying document pages, answering questions about images — it handles the job at no cost.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Connect this model's free API to my agent: https://huggingface.co/ukisai/Swift-Qwen3.8-27b — endpoint https://ukisai.com/api/swift/v1, OpenAI-compatible. Test a text query and a query with any image, show what the model responded.
This is a server-side model and cannot be run on a personal laptop. UkisAI provides a free API, so no local hardware is needed.
Open on Hugging Face▌ More finds