Current outcome
Google released Gemini 4 Argon as its next flagship long-horizon model. Its single-pass output limit jumps from 64K tokens to 1 million tokens. The model is not yet open to ordinary users; initial access is only through the Fairwind Program to trusted cybersecurity teams, with later rollout to paid API customers and Google AI Ultra users. API pricing starts at $2 per million input tokens and $10 per million output tokens, later doubling to $4 and $20. In official evaluations, Argon scores 77.9% on DeepSWE v1.1, ahead of GPT-6 Astra and Claude Opus 5.5, but still trails on FrontierSWE, Terminal-Bench 4.0 and some scientific tasks. Google emphasized cybersecurity capabilities, with trusted defender versions even removing some cyber restrictions; before formal opening, Google is still testing abuse, prompt injection and agent boundary risks.
Progress timeline
2 material updates- #01
Gemini 4 Argon: our next era of frontier intelligence
Google unveiled Gemini 4 Argon, its next frontier AI model, which the company says delivers "frontier performance in complex workflows across real-world software engineering, enterprise knowledge work like legal and finance, and cybersecurity defense," according to chief AI architect and Google DeepMind SVP Koray Kavukcuoglu. Google is limiting access to the model for now, saying it will keep gathering feedback from early testers as it iterates on guardrails before making Argon available to developers, enterprises and consumers. The announcement puts a new flagship model into the top-lab competition, but access remains restricted — developers, enterprises and consumers cannot yet use it generally, with Google citing continued work on guardrails and feedback from early testers as the reason. Google describes Argon's capabilities across coding, reasoning and multimodality, and its ability to sustain long, multi-step tasks in enterprise workflows. Hacker News commenters quoted the announcement as saying Argon agents are working on migrating C/C++ codebases to Rust across Google; independent evaluator Artificial Analysis rates Gemini 4 Argon (High) as among the leading models in intelligence and reasonably priced relative to models at a similar price point.
Source evidence: Google DeepMind Blog
- #02
Gemini 4 Argon发布：100万Token输出，首批只给可信安全团队
The new report adds key details about Gemini 4 Argon: its single-pass output limit jumps from 64K tokens to 1 million tokens; initial access is only through the Fairwind Program to trusted cybersecurity teams, with later rollout to paid API customers and Google AI Ultra users; API pricing starts at $2 per million input tokens and $10 per million output tokens, later doubling to $4 and $20; it scores 77.9% on DeepSWE v1.1, ahead of GPT-6 Astra's 74.1% and Claude Opus 5.5's 74.2%, but still trails on FrontierSWE, Terminal-Bench 4.0 and some scientific tasks; cybersecurity capabilities are emphasized, with trusted defender versions removing some cyber restrictions, while Google continues testing abuse, prompt injection and agent boundary risks.
State after update: Google released Gemini 4 Argon as its next flagship long-horizon model. Its single-pass output limit jumps from 64K tokens to 1 million tokens. The model is not yet open to ordinary users; initial access is only through the Fairwind Program to trusted cybersecurity teams, with later rollout to paid API customers and Google AI Ultra users. API pricing starts at $2 per million input tokens and $10 per million output tokens, later doubling to $4 and $20. In official evaluations, Argon scores 77.9% on DeepSWE v1.1, ahead of GPT-6 Astra and Claude Opus 5.5, but still trails on FrontierSWE, Terminal-Bench 4.0 and some scientific tasks. Google emphasized cybersecurity capabilities, with trusted defender versions even removing some cyber restrictions; before formal opening, Google is still testing abuse, prompt injection and agent boundary risks.
Source evidence: theblockbeats