Meta Muse Spark 1.3 is the latest Muse Spark model served through Meta's Model API, tuned for agentic workflows — multi-step tool use, browser operation, and long-horizon tasks — with improved coding over the previous version and a 1,048,576-token context window, suited to long documents and long conversations. It is a reasoning model: use the reasoning-effort dropdown (minimal / low / medium / high / xhigh / max, default medium; this model has no 'none' setting) and the thinking-display option to tune how deeply it thinks. Turn on web search to let the model look things up in real time and cite its sources, or upload images and documents for analysis. Use it like any other chat module: type your question or instruction and send; replies stream back. Calls go through Meta's official Model API (Standard tier; conversations are not used for model training). Best for coding, data wrangling, research, and multi-step reasoning tasks. If you need to search the web for information, please turn on the "Web Search" switch. Maximum context window (input and output combined): about 1,048,576 tokens. Maximum total upload size per request: 256MB (all files and message content combined). Accepted document formats: pdf, docx, xlsx, pptx, txt, csv (file content is provided to the model as extracted text). Accepted image formats: gif, jpeg, jpg, png, webp. Maximum number of images: 50. Maximum size per image: 47.7MB. Image dimensions: at least 200 pixels per side. Maximum image resolution: 36,000,000 pixels in total. Output content formats: Text This module supports continuous conversations. AI can make mistakes. Please verify important information.