I constructed this as we speak (with GPT-5.6-Sol xhigh) to assist take a look at Qwen 3.8 27B working in LM Studio on each my M5 MacBook Professional and an NVIDIA DGX Spark.
It gives an online UI for exercising an OpenAI-Responses-compatible chat endpoint. I’ve tried it towards LM Studio with the –cors choice and OpenRouter, and each work positive.
Conversations are persevered within the browser and may be exported as copy-pasted JSON. One enjoyable element is that it notices SVG photographs which might be being generated and progressively renders them within the chat whereas the tokens are nonetheless streaming in.
