Google DeepMind Releases Gemini 4 Argon; Meta Unveils Muse Wonder Models
New frontier models from Google and Meta intensify competition, while OpenAI adjusts pricing and venture capital continues to heavily favor AI startups.
The story
Google DeepMind publicly launched its Gemini 4 Argon model this week, making it accessible to Google AI Pro and Ultra subscribers and through its paid API starting October 13. This release positions Argon as Google's new top-tier model, designed for complex tasks in software engineering, enterprise knowledge work, and cybersecurity defense.
Meanwhile, Meta introduced its Muse Wonder and Muse Wonder Full models on October 12, marking them as the first in a new class above Muse Spark. These releases underscore the ongoing race among major labs to deploy increasingly capable models to a broader user base. Separately, OpenAI announced price reductions for its GPT-6 Sol and Luna models, aiming to make advanced AI more accessible for businesses and individual users, further heating up the competitive landscape for foundational models and their commercial applications. [2, 5, 9]
Who moved
Google DeepMind
What Changed: The lab publicly launched its Gemini 4 Argon model, making it available to Google AI Pro and Ultra subscribers and via its paid API.
Consequence: This move positions Argon as Google's most capable model for general use, broadening its accessibility for software engineering, enterprise, and cybersecurity applications. [9]
Meta
What Changed: Meta released its Muse Wonder and Muse Wonder Full models on October 12.
Consequence: These models represent a new class of AI above the Muse Spark series, indicating an expansion of Meta's model offerings for specific use cases. [2]
OpenAI
What Changed: OpenAI announced price reductions for its GPT-6 Sol and GPT-6 Luna models following its DevDay conference.
Consequence: This adjustment aims to make these advanced models more accessible and cost-effective for businesses and developers. [5]
New models
Gemini 4 Argon
Lab: Google DeepMind
What: A new frontier AI model designed for long, complex tasks.
Use: Available for Google AI Pro and Ultra subscribers and via API for software engineering, enterprise, and cybersecurity defense. [9]
Muse Wonder
Lab: Meta
What: The first of a new class of models above Muse Spark.
Use: Released to expand Meta's AI capabilities for various applications. [2]
Muse Wonder Full
Lab: Meta
What: A more capable version within the new Muse Wonder class.
Use: Available to Muse Code subscribers on $50 and $200 plans, scoring high on intelligence benchmarks. [2]
Market signals
Venture capitalists invested over $98 billion into U.S. companies in Q3 2026, with artificial intelligence accounting for 82.7% of all deal value. [3]
Implication: AI continues to attract a disproportionate share of venture capital, setting new funding records and potentially diverting capital from non-AI sectors. [3]
TypeSafe AI raised $870 million at a $7.5 billion valuation in a round led by Andreessen Horowitz. [3]
Implication: This significant funding round for an AI company specializing in machine-native models highlights strong investor confidence in enterprise AI solutions. [3]
Clueso, an AI video platform, secured a $10 million Series A funding round. [11, 19]
Implication: Smaller, specialized AI applications, particularly in agentic video creation for businesses, continue to attract investment. [11, 19]
Dapta, a startup, announced a $6 million seed investment. [11, 19]
Implication: Early-stage funding remains active for new AI ventures, indicating continued innovation at the foundational level. [11, 19]
Reporting + analyst voices: grounded via Google Search at publish time.