Observed Signal · Sep 3, 2026 · Technical Release · Source: Trending Topics (DACH/CEE Innovation & Tech) · Impact: 4/5 · Sentiment: Positive

Meta's Muse Spark 1.3 Nears Top AI Rankings

Executive Signal Summary

Meta released Muse Spark 1.3, its latest frontier AI model, scoring 61 on the Artificial Analysis Intelligence Index and 62 on max preview, tying or surpassing GPT-5.6 Sol, Grok 4.6, and Claude Opus 5, with only Anthropic's Claude Fable 5.1 and Claude Opus 5 ranking higher. Gains stem from improved agentic tasks and scientific reasoning. Pricing is aggressive at $0.55 per Intelligence Index task, while token costs remain unchanged from version 1.2. The model is available in Muse Code and via API. Meta committed to open-weight releases of Muse Spark 1.2, but 1.3 remains undecided, potentially impacting the open-source ecosystem. Additionally, OpenAI's GPT-6 Astra scores 61 but carries significantly higher API costs.

Polaris7 AgentPolaris7 Strategic Assessment
High Confidence

Meta, a major tech platform, releases a competitive AI model, impacting the AI landscape and potentially advertising technologies.

SIGNAL RADAR

Track Meta Signals & Market Shifts in Real-Time

Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your Polaris7 workspace to monitor these entities.

Start Monitoring
Verified primary sourcesAutomated briefingsInstant watchlist setup

Key Takeaways & Evidence Grounding

  • Meta released Muse Spark 1.3, scoring 61 (xhigh) and 62 (max) on the Artificial Analysis Intelligence Index, tying or surpassing GPT-5.6 Sol, Grok 4.6, and Claude Opus 5, with only Claude Fable 5.1 and Claude Opus 5 ranking higher.
  • Pricing is aggressive at $0.55 per Intelligence Index task, with token costs unchanged from 1.2 at $1.25 per million input and $4.25 per million output tokens.
  • Meta has committed to open-weight release of Muse Spark 1.2 but has not yet decided on 1.3, potentially impacting the open-source ecosystem.
  • Muse Spark 1.3 expands context window to one million tokens, accepting text, image, and video inputs.
  • OpenAI's GPT-6 Astra scores 61 but has significantly higher API costs.

Connected Companies & Entities

“Only two models from Anthropic rank higher in the overall index: Claude Fable 5.1 (max) with 66 and Claude Opus 5 (max) with 63 points.”

“OpenAI and xAI do not exceed the score of Muse Spark 1.3 in their respective strongest listed configurations.”

“The best-performing Google model in the index is Gemini 3.8 Flash (high) with 59 points.”

“Evaluations by Artificial Analysis demonstrate how far Meta has progressed.”

“According to reports by The Information, it is scheduled for release in the fall.”

Ontology Mapping & Concepts

Primary Source Grounding & Direct Attribution
Direct Origin Attribution
Primary Reporting: Trending Topics (DACH/CEE Innovation & Tech)•Published: Sep 3, 2026
Original Coverage Title: “Muse Spark 1.3: Meta ist zurück an der Spitze und könnte bald das beste Open-Weight-Modell bieten”

Related Market Signals & Shifts

Recent verified developments and strategic activity across this market segment.

InfrastructureSep 3, 2026

Meta Launches Muse Spark 1.3 to Rival GPT-5.6 Sol

Meta has released Muse Spark 1.3, the latest iteration of its AI model family, designed for longer agentic workflows, complex coding tasks, and reliable processing of extensive instructions. The model improves collaboration by maintaining context in long chats, distinguishing multiple tasks, and asking for clarification. It also enhances safety against prompt injection and irreversible actions. Reportedly, Muse Spark 1.3 uses about 20% fewer tool calls and 25% fewer tokens than its predecessor, with a more compact coding style. In Meta's benchmarks, it outperforms rivals GPT-5.6 Sol and Opus 5 on the long-context MRCR test (98.5 vs 91.5 points) and several coding benchmarks, though it trails in others like DeepSearchQA. The model is available in Muse Code and via the Meta Model API, with a Max-Reasoning mode and an Open-Weights version announced.

Read assessment
Large Language Models (LLM) & AIJul 10, 2026

Meta unveils Muse Spark 1.1 agent model

Meta released Muse Spark 1.1, an agent-focused multimodal model designed to plan tasks, coordinate parallel AI sub‑agents, use external tools and execute end‑to‑end workflows. Muse Spark 1.1 supports images, video and audio, claims a context window up to one million tokens, and emphasizes visual‑to‑code generation and agentic control of desktop/web interfaces. Meta published internal benchmarks placing Muse Spark 1.1 competitive with current frontier models (GPT‑5.5, Claude Opus 4.8, Google Gemini 3.1 Pro) for agents and coding tasks. Meta also published API pricing (USD 1.25 per 1M input tokens; USD 4.25 per 1M output tokens) and offers $20 trial credit for new accounts. Demos shown include automating a Facebook Marketplace listing and debugging/code migration workflows, and Meta says developers and researchers already use the model for daily development and research tasks.

Read assessment
Large Language Models (LLM) & AIApr 9, 2026

Meta AI App Hits No. 5 After Muse Spark Launch

Meta launched Muse Spark, a new multimodal AI model, and updated its Meta AI app and website. Appfigures data show the Meta AI app jumped from No. 57 to No. 5 on the U.S. App Store immediately after the Muse Spark release, suggesting a large increase in installs. Appfigures reports 60.5 million total installs across App Store and Google Play (25 million in 2026) and a 138% download increase over the app’s first five months. Muse Spark accepts voice, text and image inputs, can spin up multiple subagents, and is designed to help with tasks such as health education, complex reasoning, and visual coding. Meta plans to roll the model out to WhatsApp, Instagram, Facebook, Messenger and its AI glasses in the coming weeks. Meta positions Muse Spark as an upgrade to Llama 4 and a step toward competing with OpenAI, Anthropic and Google.

Read assessment
Executive Briefing & Watchlist

Receive all future signals for Meta & competitors in your briefing

Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.

Daily primary source scansCurated executive briefingsUnsubscribe anytime