Meta Muse Spark 1.1 Evaluation Released
ArtificialAnlys · x · 2026-07-11
Artificial Analysis reports that Meta's Muse Spark 1.1 scored 51 on the Intelligence Index, an 8-point increase over version 1.0, outperforming peers in both cost and token efficiency.
Key findings include:
- Significant gains in Scientific Reasoning, coding, and knowledge tasks, though it still trails stronger models in agentic knowledge work.
- Scored 45% on Humanity's Last Exam, closely trailing Claude Opus 4.8 (46%) and beating GPT-5.5 (44%) and Grok 4.5 (40%).
- Consumed only 94M output tokens to run the Intelligence Index, costing an estimated $0.26 per task—lower than GLM-5.2 and GPT-5.4.
- Other details: 1M context window, $1.25/$4.25 per million input/output tokens, and available directly via Meta's first-party API at launch.
More from Models
- Rerunning Argon on Antigravity Harness lifts it past Astra, near Opus 5.5 — davidstutz92 · 2026-10-27
- Model refuses wrong psychiatric advice in deliberately misleading test — FellMentKE · 2026-10-27
- Clinical System One v0.2: top clinical AI drops from 97% to 65% on harder cases — MaziyarPanahi · 2026-10-27
- ChatGPT keeps telling people 404 Media is tech support for big corporations — 404 Media · 2026-10-26
- Claude's /show-me visualizes compressor auto makeup gain like a pro audio teacher — teropa · 2026-10-26
- ChatGPT keeps telling people 404 Media can fix their tech support problems — 404 Media · 2026-10-26