Meta Muse Wonder and Muse Wonder Full
Meta's new frontier models from Meta Superintelligence Labs: Muse Wonder (1M-token context, API + Muse Code) and the stronger Wonder Full (Muse Code subscribers only), with Active Memory and Active Focus for long-running agent tasks.
Why it matters
Meta's first genuinely frontier-competitive model in a while: Wonder Full ties Claude Sonnet 5.5 on the Artificial Analysis Intelligence Index at 56, and the Active Focus mechanism targets the exact failure mode of agents that quit at the first error.
What you could build with it
A team building a long-horizon research assistant could prototype on Muse Wonder's 1M-token context and cross-conversation memory to test whether persistent memory actually cuts repeated clarification questions, measuring task completion per dollar against their current model.
Does it hold up?
Too early to judge - launched Oct 12; independent benchmarks exist but production usage reports are not out yet.
Built with Meta Muse Wonder and Muse Wonder Full
- Meta Launched Muse Wonder. Let's Build a Tiny Active Focus Loop in TypeScript.dev.to · Hands-on tutorial modeling Active Focus as a small TypeScript agent loop that switches approaches, asks for clarification, and checks in before risky actions.
- Muse Wonder: Meta's Frontier Model, Benchmarks and Priceblog · Independent writeup covering benchmarks, pricing and availability limits of both Wonder models.
- Meta launches Muse Wonder models, hitting top AI benchmark scoresnews · News coverage with benchmark detail: Wonder Full scores 56 and Wonder 52 on the Artificial Analysis Intelligence Index.
Learn more
First spotted on article: source.
More AI models
OpenAI publishes 722 AI-generated math manuscripts from an unreleased modelOpenAI released 722 mathematical manuscripts in 372 result families to GitHub under Apache-2.0, produced by an…models · JEV 0.74Mistral Large 4 (Le Chonk): 1.05T-param open-weight MoEMistral's new flagship: a 1.05T-parameter mixture-of-experts multimodal model with 49B active params and 1M-token…models · JEV 0.72ElevenLabs Eleven v4 + v4 Turbo: new emotive TTS architecture with a 100ms real-time variantElevenLabs released Eleven v4 and Eleven v4 Turbo on September 28: a new text-to-speech architecture with inline…models · JEV 0.72VoiceStudio: open-source, fully-local ElevenLabs alternative with voice cloning, dubbing and dictation in 646 languagesVoiceStudio is an open-source, fully-local voice AI studio: voice cloning from a 3-second sample, voice design, video…models · JEV 0.71JetBrains Mellum 2.1: 12B MoE coding model trained for agentsJetBrains released Mellum 2.1, a 12B MoE model (2.5B active parameters, 128K context) under Apache 2.0, trained mainly…models · JEV 0.69DeepSeek V4-Flash official API: public beta with upgraded agent capabilities, Responses API and Codex supportDeepSeek's official V4-Flash is now a public beta API with massively upgraded agent capabilities - benchmark scores…models · JEV 0.69
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.