{
  "id": 21021496,
  "title": "Google launches Gemini 4 Neon taking on Claude Opus 5.5 and GPT-6.2 Sol",
  "url": "https://urgent.news/2026/11/21/google-launches-gemini-4-neon-taking-on-claude-opus-5-5-and-gpt-6-2",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-21T08:33:51.000Z",
  "source": {
    "name": "The Hindu - Sci-Tech",
    "slug": "the-hindu-sci-tech",
    "url": "https://www.thehindu.com/sci-tech/technology/google-launches-gemini-4-neon-taking-on-claude-opus-55-and-gpt-62-sol/article71727599.ece"
  },
  "original_language": "en",
  "account": "Google has introduced an advanced AI model, Gemini 4 Neon, designed to excel in complex software engineering, knowledge work, and science workflows. The company claims that Gemini 4 Neon outperformed comparable models, such as Anthropic’s Claude Opus 5.5 and OpenAI’s GPT-6.2 Sol, on several benchmarks.\n\nPriced at $3 per million input tokens and $15 per million output tokens, Gemini 4 Neon features a million-token context window. This model is specifically tailored for long-running agentic workflows, ensuring that tasks are completed while maintaining cost efficiency.\n\nGoogle asserts that users of the Gemini app will experience enhanced writing capabilities and improved performance on knowledge work, research, and challenging tasks. The company highlights the model's ability to handle tedious tasks, like analyzing billing statements to identify unusual charges or cross-checking documents to negotiate refunds.\n\nGoogle reassures that Gemini 4 Neon maintains the same carbon-level safeguards against severe misuse and prompt injection as its predecessor, alongside enhanced model alignment. The new model can be accessed via the Gemini API in Google AI Studio, through the Gemini for business API and model dropdown, as well as on Google Antigravity for Google AI Pro and Ultra subscribers, and within the Gemini app for Pro and Ultra subscribers.\n\nAccording to Koray Kavukcuoglu, SVP of Google DeepMind and Chief AI Architect at Google, \"Neon is built for complex coding, analytical reasoning, and production-grade enterprise workflows. While Carbon tackles your hardest, long-horizon projects, and Hydrogen handles fast everyday execution, Neon balances depth and efficiency. Neon delivers frontier-level reasoning for complex, high-frequency workflows faster and more cost efficiently.\"",
  "summary": "Google said the new model was built for complex coding, analytical reasoning, and enterprise workflows",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 3,
    "also_reported_by": [
      {
        "outlet": "The Indian Express",
        "title": "Google debuts Gemini 4 Neon, an AI model aimed at complex tasks for enterprises",
        "url": "https://urgent.news/2026/11/19/google-debuts-gemini-4-neon-an-ai-model-aimed-at-complex-tasks-for",
        "published": "2026-11-19T10:02:51.000Z"
      },
      {
        "outlet": "Google Research",
        "title": "How Google Duplex evolved to also call businesses for you in Gemini",
        "url": "https://urgent.news/2026/11/20/how-google-duplex-evolved-to-also-call-businesses-for-you-in-gemini",
        "published": "2026-11-20T23:30:30.000Z"
      }
    ]
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}