by Llama
Muse Spark 1.1 is a multimodal reasoning model from Meta, built for agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text, with a 1M-token context window.
| Input | $1.38 / 1M tokens |
| Output | $4.67 / 1M tokens |
| Context window | 1,000,000 tokens |
| Modalities | text, image, audio, video |
| Features | thinking, tools |
Use muse-spark-1.1 via the AIHubMix unified API — one interface for every major LLM.