Google Unveils Gemini 4 Argon With 1M-Token Output Limit
Google announced Gemini 4 Argon, a new frontier model with a 1M-token output limit and top scores on coding and multimodal benchmarks.
Google announced Gemini 4 Argon, a new frontier model that introduces a new naming scheme for the company's AI lineup. The model raises the output token limit to 1M tokens, up from 64K.
Google reports a DeepSWE v1.1 score of 77.9% for Gemini 4 Argon. On CWE-bench v1, the model ties for first place with a score of 68%.
On AutomationBench, Gemini 4 Argon ranks first with a score of 51.3%. On LVBench, it achieves a state-of-the-art score of 91.7%.
Gemini 4 Argon will roll out soon, starting with Google AI Ultra subscribers and paid API customers. Google will also make the model available without cyber guardrails to trusted defenders and internal Google teams.
Background
Gemini 4 Argon follows the cancellation of Gemini 3.5 Pro and a focus on Gemini 3.8 Flash. The model had been made available to trusted testers and cyber defenders via the Fairwind Program, as reported by 9to5google.com.
Quick answers
What is Gemini 4 Argon's output token limit?
Gemini 4 Argon has an output token limit of 1M tokens, up from 64K.
Who gets Gemini 4 Argon first?
It will roll out soon, starting with Google AI Ultra subscribers and paid API customers.
What benchmarks did Google report for Gemini 4 Argon?
Google reports 77.9% on DeepSWE v1.1, 68% on CWE-bench v1, 51.3% on AutomationBench, and 91.7% on LVBench.