Today for AI
HOT RADAR
llmHEAT 32.9°

AI CLUSTERED EVENT · 10/13/2026

Meta Launches Muse Wonder: Leveraging Prometheus Data Center and Active Memory to Surpass Claude Opus 5.5 in SWE-Marathon with Score of 57.5

17 reports archived3 independent sourcesupdated 10/15/2026, 04:19:17
Synthesis & Latest Updates
3 Sources Cross-Validated

Meta Superintelligence Labs (MSL) has officially launched Muse Wonder, its first closed-source frontier model, alongside the enhanced Full version. Trained on the gigawatt-scale Prometheus data center in Ohio, this release marks Meta's return to the top tier of AI development just 14 months after establishing MSL. The core innovation lies in introducing "Active Memory" and "Active Focus" mechanisms, which significantly enhance the agent's ability to persist information across sessions and autonomously navigate long-horizon complex tasks, shifting from pure reasoning to autonomous action.

Benchmark data reveals that Muse Wonder Full scores 56 on the Artificial Analysis Intelligence Index v4.3, an 8-point increase over its predecessor. Its agentic capabilities have surged, with Terminal-Bench 4.0 scores jumping from 33% to 57% and GDPval-AA v2.1 Elo ratings rising by 176 points. Notably, it achieved a score of 57.5 in the SWE-Marathon test, substantially outperforming Claude Opus 5.5's 46.3, and scored 79.0 in DeepSWE coding benchmarks, demonstrating superior performance in long-term code generation and legal verticals.

Commercially, Meta employs a tiered strategy: the standard version is available via API ($4/M input, $20/M output), while the full-capability Wonder Full is restricted to Muse Code subscribers. This approach aims to build a personal and professional agent ecosystem and capture high-value B2B market share. Despite initial billing glitches, this launch establishes Meta as a direct competitor to Anthropic and OpenAI, offering developers a powerful solution for long-horizon tasks.

LATEST/Muse Wonder Standard is now available via API with 1M token context and multimodal support, while Wonder Full remains exclusive to Muse Code subscribers. General user access timelines are unclear due to conflicting official statements. Initial deployment faced quota tracking bugs, highlighting stability challenges, though Zuckerberg claims the model matches or exceeds GPT-6 Astra and Claude Opus 5.5 in specific benchmarks.

HEAT TRENDHourly heat curve

62 fully observed hours
Now
32.9
Peak
187.810:00
24h change
-73%

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

TIMELINECoverage timeline

Total 16 reports · Latest first
  1. Arena.ai (@arena)T1·68 pts
    • Muse Wonder (Max) ranks #12 in Agent Arena, improving by 4 places and 2.8 points over Spark 1.3.
    • Median cost per task is $1.13; while higher than Spark's $0.54, it is 25% cheaper than similarly ranked GPT-5.6 Sol and 67% cheaper than Claude Fable 5.
    • The model shows significant performance gains over its predecessor in Chat, Work, and Code categories.
  3. InfoQ 中文T2·86 pts
    • Meta releases the Muse Wonder series, positioning it as a personal and professional-grade AI Agent capable of autonomous action rather than just reasoning.
    • Key innovations include 'Active Memory' for cross-session information persistence and 'Active Focus' for dynamic adjustment in long-horizon tasks.
    • Clear commercialization path: Base version uses token-based pricing ($4/M input, $20/M output), while the advanced Wonder Full version is subscription-only.
  4. Latent SpaceT1·78 pts
    • Meta MSL released its second-generation model, Muse Wonder (codename 'Watermelon'), aiming to establish a multi-tier product matrix similar to OpenAI/Anthropic.
    • Muse Wonder Full outperformed competitors on some benchmarks; while exact parameter counts are undisclosed, third-party evals indicate frontier-level competitiveness.
    • The lab achieved a Claude 5.5-tier equivalent model in just 14 months from inception, demonstrating high R&D iteration efficiency.
  5. IT之家 · IT业界T2·78 pts
    • Muse Wonder adopts a closed-source strategy, selling its most powerful version directly to enterprises to compete with Anthropic and OpenAI for high-value B2B contracts.
    • Official claims indicate superior performance over Claude Opus 5.5 and GPT-6 Astra in complex tasks like contract interpretation and financial data organization.
    • The new model will be integrated into the Muse personal assistant app, which currently tops the US App Store charts, aiming to strengthen C-end user retention.
    • The launch coincided with technical bugs (e.g., quota tracking errors in Meta Code subscriptions), highlighting stability challenges during early large-scale deployment.
  7. IT之家 · 智能时代T2·78 pts
    • The Muse Wonder series demonstrates strong performance in reasoning, writing, and coding benchmarks, approaching Anthropic's top-tier scores.
    • This is Meta's first product explicitly positioned as a 'Frontier Model,' requiring high compute power and developed under Alexandr Wang's leadership.
    • The Muse Wonder Full version is exclusive to Muse Code subscribers, indicating Meta's strategy to expand into B2B and enterprise services via premium coding tools.
  8. The DecoderT2·86 pts
    • Muse Wonder Full scored 56 on the Artificial Analysis Intelligence Index v4.3, with the standard version at 52, significantly surpassing the predecessor Muse Spark 1.3 (48).
    • Performance on Terminal-Bench 4.0, which measures autonomous complex task handling, jumped dramatically from 33% to 57%, indicating top-tier agentic capabilities.
    • The new models introduce an 'Active Memory' feature for retaining key information and reduce output token usage by approximately 10% compared to predecessors, enhancing inference efficiency.
  9. Rohan Paul (@rohanpaul_ai)T2·78 pts
    • Muse Wonder scores 57.5 on SWE-Marathon, creating a significant gap compared to Claude Opus 5.5's 46.3 and GPT-6 Astra's performance.
    • SWE-Marathon requires AI to independently build full software projects like a C compiler or Slack clone within 2-10 hours, counting success only if all hidden checks pass.
    • Meta positions Muse Wonder as its largest and most capable model, emphasizing competitiveness in both personal and professional agent scenarios.
  10. Arena.ai (@arena)T1·68 pts
    • The Muse Wonder series represents Meta's most capable models yet, specifically engineered for long-horizon agentic tasks.
    • Models are accessible via Meta Model API and Muse Code, with Muse Wonder Full restricted to subscribers.
    • The series has joined Agent Arena, where performance is measured against millions of real-world tasks using causal tracing, alongside support for various code generation modalities.
  11. TechCrunch AIT2·86 pts
    • Muse Wonder represents Meta's first dedicated attempt at a closed-source frontier model, benchmarking against GPT-6 and Claude.
    • Tiered access strategy: Wonder is available via API, while Wonder Full is exclusive to Muse Code subscribers, signaling a commercial moat.
    • Infrastructure milestone: Training leverages the newly operational Prometheus data center, highlighting Meta's compute capacity expansion.
  12. 宝玉 (@dotey)T2·86 pts
    • Muse Wonder comes in Standard (speed-focused) and Full versions, trained on Meta's gigawatt-scale Prometheus data center.
    • Key strengths lie in agentic and long-horizon tasks, outperforming rivals in Personal Superintelligence Index (59.1 vs 55.2) and DeepSWE (79.0 vs 74.2).
    • Access is tiered: developers can use APIs/Muse Code today, Full version is subscription-only, while general user rollout timelines remain ambiguous and inconsistent across official channels.
  13. Artificial Analysis (@ArtificialAnlys)T1·78 pts
    • Muse Wonder Full achieves an Intelligence Index score of 56, an 8-point improvement over its predecessor, outperforming Muse Spark 1.3 across all ten evaluations.
    • Significant gains in agentic performance: Terminal-Bench 4.0 scores jumped from 33% to 57%, and GDPval-AA v2.1 Elo increased by 176 points.
    • Supports text, image, video, and audio inputs; Muse Wonder is available via API, while Muse Wonder Full is currently exclusive to Muse Code users.
  14. Rohan Paul (@rohanpaul_ai)T2·86 pts
    • Muse Wonder flagship beats Claude Opus 5.5 on 8 of 14 benchmarks and leads GPT-6 Astra in long-running tasks.
    • Scores 57.5 on SWE-Marathon (vs. 46.3 for competitor) and 79.0 on DeepSWE 1.1, demonstrating superior long-horizon engineering capabilities.
    • Achieves 13.1 on Harvey’s legal agent benchmark, more than triple Claude Opus 5.5’s score, highlighting robustness in hallucination-free task completion.
  15. Yuchen Jin (@Yuchenj_UW)T1·68 pts
    • Muse Wonder surpasses competitors Opus 5.5 and Astra on SWE-Marathon and DeepSWE v1.1 coding benchmarks.
    • Artificial Analysis Intelligence Index shows a minimal gap (only 2 points) behind top-tier model Opus 5.5.
    • Strong community demand for Meta to open-source Muse Wonder or provide lightweight alternatives like Spark 1.3.
  16. Artificial AnalysisT1·86 pts
    • Muse Wonder Full achieves an Intelligence Index score of 56, an 8-point gain over its predecessor, improving across all ten evaluations.
    • Significant agentic capability boost: Terminal-Bench 4.0 scores rose from 33% to 57%, and GDPval-AA v2.1 improved by 176 Elo points.
    • Clear tiering strategy: Muse Wonder is available via API, while Muse Wonder Full is currently exclusive to Muse Code users.