Meta open-weights Muse Spark 1.2: video-to-working-code demos and broad multimodal gains
qinzytech · x · 2026-10-16
Meta released Muse Spark 1.2, a coding-focused update to Muse Spark 1.1, with model weights freely available on Hugging Face and a new evaluation report.
- Visual coding: given an image or video, the model translates layout, hierarchy and style into working web pages and games, re-inspecting its own artifacts in a self-improvement loop; demos include a single-page interactive flipbook with generated typography, page-turn animations and a cursor-tracked magnifier.
- Audio-visual understanding: strengthened for video-heavy enterprise workflows.
- Tool use: Meta says multimodal gains are most pronounced when the model can use tools, letting it inspect visual inputs and fold findings into reasoning.
- A featured demo turned a terminal-uploaded video into a functional home-booking page, framed as multimodal reasoning grounded in action.
Related event: Meta Open-Sources Muse Spark 1.2 and Previews WildArtifactBench(2 posts)→
More from Models
- OpenAI's Decisions API hits GA: near real-time routing up to 10x faster than GPT-6 Luna — OpenAIDevs · 2026-10-23
- DeepLearning.AI weekly: GPT-6.1 hits 97.9% byte parity on COBOL migration — DeepLearningAI · 2026-10-23
- Marker's lite mode scores 75.1 vs 42.3 for Azure Read on OmniParseBench — VikParuchuri · 2026-10-23
- Datalab launches lite OCR at $0.50 per 1,000 pages, 20-30x cheaper than Azure, Google and AWS layout APIs — VikParuchuri · 2026-10-23
- LiveNerf's 29-day data finds no nerfing of Claude Opus 5.5 — TheOnlyVibemaster · 2026-10-23
- PHP Foundation reveals Anthropic's Claude cyber model scanned PHP, validated with OpenAI's Daybreak Red — saibharadwaj · 2026-10-23