

Muse Code
#6 in Coding-AgentenMeta · siet 5. August 2026 (Beta) · 135× · tolest 28. Okt. 2026
Muse Code is a terminal-based Coding Agent from Meta (Meta Superintelligence Labs), available as Beta for macOS and Linux since August 5, 2026. It is powered by the new Muse Spark 1.2 model and can plan complete software development tasks across large repositories, write code, and validate results, supported by persistent background sub-agents and a local event log for crash recovery. The product competes directly with Claude Code and OpenAI Codex and positions itself through a significantly cheaper pricing model, including a "Contributor" tier with a discount in exchange for consenting to data use for model training. There is no dedicated app or IDE integration; usage occurs exclusively via the command line.
Features
| Base Model | Muse Spark 1.2 (co-trained with the Muse Code harness, 1-million-token context window) |
| License | Proprietary, closed-source/closed-weight; no public weights, usable only via Meta Model API/Muse Code binary |
| Platform | macOS and Linux, one-line terminal install command; no native Windows support, no desktop app |
| Price | Standard: $1.25/M input tokens, $4.25/M output tokens (cached input $0.15); Contributor tier: $0.10/M input, $0.20/M output (cached input $0.002) in exchange for data training rights |
| Release Date | August 5, 2026 (beta) |
| Interface (IDE/CLI/Web) | CLI (terminal) only; no dedicated app, no VS Code or JetBrains plugin |
| SWE-bench Verified (%) | 86,6 % (Vals AI, Harness Mini-SWE-agent; Meta-Eigenangabe nicht gefunden) |
Mehr Produkten in disse Kategorie: Coding-Agenten
Belege (60)
Company Analysis: Meta
Data supports a balanced stance. Positives: strong operating momentum in the core ad business (Q3 2026 revenue +26% YoY; advertising $63.1B; 3.64B DAP) indicates Meta’s platform remains highly competitive for advertisers. (investor.atmeta.com) Offsets: profitability and cash conversion weakened materially in the quarter (20% operating margin; negative Q3 free cash flow driven by capex), and management’s disclosed 2026 capex range ($130B–$145B) plus large infrastructure commitments imply continued pressure on near-term FCF and margin stability. (investor.atmeta.com) Additionally, EU regulatory scrutiny (DSA/DMA-related actions) remains an active, potentially economically relevant risk factor. (digital-strategy.ec.europa.eu) Given these cross-currents and the user’s “current price unknown” constraint (limiting precise EUR valuation work), HOLD is the most data-consistent recommendation until there is clearer evidence of (1) capex normalization or (2) margin/FCF re-acceleration while maintaining current revenue growth.
Summary
Meta Platforms, Inc. (“Meta”) monetizes attention and intent primarily through performance and brand advertising across its Family of Apps (Facebook, Instagram, Messenger, WhatsApp) and, to a much smaller extent, through Reality Labs (AR/VR hardware, software, and content). The core competency is large-scale consumer engagement paired with ad targeting/measurement, increasingly supported by AI-driven ranking, creative tools, and advertiser automation. Meta’s economic engine remains advertising: in Q3 2026, advertising revenue was $63.1B out of total revenue of $64.6B (+26% YoY). (investor.atmeta.com) Market position is anchored by global reach and engagement: Family daily active people (DAP) averaged 3.64B in September 2026 (+3% YoY), supporting a durable demand-side platform for advertisers and a large supply of ad impressions. (investor.atmeta.com) Competitive advantages include (1) scale and cross-app data/identity graph, (2) AI ranking and ad delivery efficiency, and (3) a broad advertiser base (SMBs to large brands) with measurable ROI. Recent developments (last ~90 days) are dominated by: (a) Q3 2026 results showing strong top-line growth but materially lower operating margin due to cost growth and large investment intensity; (b) continued heavy AI infrastructure commitments and capex guidance; and (c) elevated regulatory scrutiny in Europe. In Q3 2026, Meta reported revenue of $64.6B (+26% YoY) and operating income of $13.0B (operating margin 20% vs 40% in Q3 2025). Segmentally, Family of Apps generated $64.2B revenue and $17.1B operating income, while Reality Labs produced $443M revenue and a $(4.13)B operating loss. (investor.atmeta.com) Free cash flow was negative in Q3 2026 (FCF $(2.1)B) due to very high property and equipment purchases. (investor.atmeta.com) On investment intensity, Meta disclosed anticipated 2026 capital expenditures of approximately $130B–$145B to support AI efforts and the core business, alongside very large multi-year infrastructure lease and cloud commitments. (sec.gov) This is a key swing factor for near-term margins and cash generation. Regulatory and legal risk remains a meaningful overhang in the EU: the European Commission stated (July 10, 2026) a preliminary finding that certain “addictive design” features may breach the Digital Services Act, and recent reporting indicates additional scrutiny around business terms and data practices. (digital-strategy.ec.europa.eu) Outlook (short- to medium-term): operationally, Meta’s revenue trajectory is currently strong, but the investment cycle (AI infrastructure + Reality Labs losses) is compressing margins and increasing cash flow volatility. Consensus/market data sources indicate investors still broadly expect earnings growth and maintain positive ratings, but the key debate is whether incremental AI spend translates into sustained ad yield improvements and/or new monetization, fast enough to offset cost growth. (stockanalysis.com)
Key Takeaways
- Q3 2026 revenue growth remained strong (+26% YoY), with advertising still the dominant driver ($63.1B of $64.6B). (investor.atmeta.com)
- Profitability compressed sharply: Q3 2026 operating margin fell to 20% (from 40% in Q3 2025), reflecting materially higher costs and investment intensity. (investor.atmeta.com)
- Reality Labs remains a large drag: Q3 2026 RL operating loss was about $(4.13)B on $443M revenue. (investor.atmeta.com)
- Capex is the central near-term financial variable: Meta guided to ~$130B–$145B capex in 2026, implying continued pressure on free cash flow and returns if monetization lags. (sec.gov)
- EU regulatory scrutiny is active and could affect product design, engagement mechanics, and ad personalization economics (DSA/DMA-related actions and investigations). (digital-strategy.ec.europa.eu)
Action Ideas
Accumulate on the view that Meta’s current operating performance (26% YoY revenue growth in Q3 2026) and scale (3.64B DAP) can sustain above-market ad growth, while AI-driven ranking and advertiser tooling continue to improve ROI and pricing power. The key underwriting point is that the Family of Apps segment still generates substantial operating income ($17.1B in Q3 2026) even during a heavy investment phase, providing capacity to fund AI capex while maintaining strategic flexibility. ([investor.atmeta.com](https://investor.atmeta.com/investor-news/press-release-details/2026/Meta-Reports-Third-Quarter-2026-Results/default.aspx))
Horizon: 18 mo.
Maintain exposure but avoid adding until there is clearer evidence that AI infrastructure spend is stabilizing relative to revenue and that operating margin can re-expand from the current compressed level. This stance is supported by the combination of strong top-line momentum and engagement, offset by (1) negative Q3 free cash flow driven by capex and (2) ongoing Reality Labs losses. ([investor.atmeta.com](https://investor.atmeta.com/investor-news/press-release-details/2026/Meta-Reports-Third-Quarter-2026-Results/default.aspx))
Horizon: 12 mo.
Reduce exposure if your mandate prioritizes near-term cash generation and margin stability: Q3 2026 showed a step-down in operating margin (20%) and negative quarterly free cash flow (about $(2.1)B), while management’s disclosed 2026 capex range ($130B–$145B) implies continued heavy cash absorption. This action is most appropriate where valuation discipline is tied to near-term FCF yield rather than longer-duration AI optionality. ([investor.atmeta.com](https://investor.atmeta.com/investor-news/press-release-details/2026/Meta-Reports-Third-Quarter-2026-Results/default.aspx))
Horizon: 6 mo.
Contrarian Insights
- • The market focus on AI capex as purely margin-negative may underweight the possibility that Meta’s AI spend is already embedded in measurable ad revenue acceleration: Q3 2026 advertising revenue grew 26% YoY, suggesting the platform is still monetizing effectively even during peak infrastructure build-out. The contrarian angle is that the investment cycle may be less dilutive to long-run unit economics than headline capex implies, if it sustains higher ad yield and retention. (investor.atmeta.com)
- • Consensus often treats EU regulation as a steady, manageable headwind; a contrarian view is that EU enforcement could become a more direct product-design constraint (e.g., engagement mechanics and recommender systems) with second-order impacts on time spent and ad inventory. The July 10, 2026 preliminary DSA finding on “addictive design” increases the probability of mandated UX changes, which could be more economically material than incremental fines. (digital-strategy.ec.europa.eu)
Sources (7)
- https://investor.atmeta.com/investor-news/press-release-details/2026/Meta-Reports-Third-Quarter-2026-Results/default.aspx
- https://www.sec.gov/Archives/edgar/data/1326801/000162828026066773/0001628280-26-066773-index.htm
- https://www.sec.gov/Archives/edgar/data/1326801/000162828026050705/meta-20260630.htm
- https://digital-strategy.ec.europa.eu/en/news/commission-preliminarily-finds-addictive-design-instagram-and-facebook-breach-digital-services-act
- https://apnews.com/article/norway-media-meta-competition-european-union-united-kingdom-c15442c8e505e50e45ce3fd7bbaae1ba
- https://stockanalysis.com/stocks/meta/forecast/
- https://www.spglobal.com/market-intelligence/en/news-insights/research/2026/07/microsoft-and-meta-earnings-previews