Meta's first image model Muse Image debuts #4 in image editing at just $0.01 per image
ArtificialAnlys · x · 2026-09-04
Artificial Analysis released evaluations of Muse Image, the first image model from Meta Superintelligence Labs:
- Rankings: #4 on the Image Editing leaderboard (behind Microsoft's MAI-Image-2.6-Preview and OpenAI's GPT Image 2 (high), narrowly behind Google's Nano Banana 2); #5 in Text to Image. Launched in Meta AI in July, now available to developers via the Meta Model API.
- Positioning: Meta calls it an agentic image model — it invokes search and coding tools, self-refines its generations, and composes from multiple references.
- Pricing: $0.01 per image ($10/1k), placing it on the quality-vs-price Pareto frontier; the next frontier model, Nano Banana 2, costs 6x more per image.
- Strengths: near-frontier on Lighting and Reasoning for text-to-image; at the category frontier on Scene & Style Edits and Identity Preserving Edits; best use cases are social media/creator content and animation/gaming assets.
More from Multimodal
- Voxtral 3 ships a LiveKit-based voice agent SDK streaming listen-think-speak — MistralAI · 2026-10-21
- Voxtral Mini TTS 3 hits SOTA expressiveness in 35+ languages with tag-based control — MistralAI · 2026-10-21
- Mistral launches Voxtral 3 speech models and new Voice Agents API — MistralAI · 2026-10-21
- Open-sourced pixel art normals generator ships as Rust CLI and wasm library — HugoDuprez · 2026-10-20
- Fibo 2.0 is strongest for ecommerce and creator content, weakest at UIUX design — ArtificialAnlys · 2026-10-20
- Fibo 2.0 leads old Fibo on all 9 capabilities, best at reasoning and anatomy — ArtificialAnlys · 2026-10-20