Enterprise budgets under strain as Google Gemini 4 Argon intro pricing expires in November
The new AI model offers strong coding performance, but analysts warn that introductory rates will end soon, complicating enterprise budgets.
Key points
- Google launched Gemini 4 Argon, a flagship model for complex coding and enterprise tasks.
- Introductory pricing of $2 per million input tokens doubles to $4 after 30 November.
- Analysts advise budgeting for the higher rate immediately due to long procurement cycles.
Google has released Gemini 4 Argon, a new artificial intelligence model designed for complex software development and enterprise workloads, with an introductory price that will double after 30 November. The launch targets organisations seeking advanced reasoning capabilities for long-term projects, according to InfoWorld. The company positions the tool as capable of handling deep, structural tasks that require sustained precision over extended periods.
The model is available via the Gemini API in AI Studio and on Google’s Gemini platform, as well as to Google AI Ultra subscribers in the Antigravity environment. Google engineers reportedly used Argon to refactor six million lines of code in Chromium, completing the work in weeks rather than months. Teams in Antigravity saw a 35% rise in agentic code submissions and a near-third reduction in rollbacks, the company said.
What we know
Pricing during the introductory period is set at $2 per million input tokens and $10 per million output tokens, with cached inputs discounted by 95%. A footnote in Google’s blog post states these rates rise to $4 and $20 respectively after 30 November. The model shares a one-million-token limit between input and output within the Gemini API.
Sanchit Vir Gogia, chief analyst at Greyhound Research, told Computerworld that chief information officers should budget for the higher rate from day one. He noted that enterprise security reviews and procurement cycles rarely finish within the short discount window, meaning business cases built on launch rates will likely need rewriting.
Gogia added that Google bills reasoning tokens at the output rate, making per-token prices a poor indicator of total agent costs. Sakshi Grover, research director at IDC, said agentic workloads multiply token usage through planning loops and tool calls. She cited an IDC survey showing 72% of enterprises experienced AI agent spending overruns in the past year.
Defender's checklist
- Review current AI usage metrics to establish a baseline before adopting new models.
- Ensure procurement teams account for post-discount pricing in annual budgets.
- Verify that security reviews are scheduled to complete before any discount periods end.
- Monitor token consumption closely, as reasoning tasks may incur higher output costs.
The background in plain language
Organisations adopting AI agents must understand that cost structures differ from traditional software licensing. Agents often perform multiple steps, such as planning and tool calling, which consume additional processing tokens. Teams usually find that per-token prices do not reflect the total cost of a completed task. Check whether your evaluation processes align with vendor discount windows to avoid unexpected budget increases.
Background: AI agents
AI agents are the software entities that perform specific tasks using instructions from large language models. Agentic AI is the architectural approach that grants these agents autonomy to plan, reason and act. You use agents as tools within agentic AI systems to achieve complex, multi-step outcomes without manual oversight at every stage.
Read the full guide: AI Agents vs Agentic AI: The Distinction That Matters
What to do and how to stay safe: Google
- Audit existing AI contracts to identify any introductory pricing clauses that may expire soon.
- Engage finance and security teams early to align procurement timelines with vendor promotions.
- Track actual token usage in pilot projects to estimate true costs beyond per-token rates.
- Review internal security policies to ensure they accommodate the specific requirements of agentic workflows.
Step-by-step guide: AI Governance for Small Teams: Practical Controls and Cost
General security guidance from the TechWire Beat newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
When does the introductory pricing for Gemini 4 Argon end?
The introductory pricing ends after 30 November, at which point rates double.
How much does Gemini 4 Argon cost during the introductory period?
It costs $2 per million input tokens and $10 per million output tokens, with cached inputs discounted by 95%.
What are the pricing rates after the introductory period?
Prices rise to $4 per million input tokens and $20 per million output tokens after 30 November.
