Wirestack
Gemini 4 Argon is generally available on October 13, 2026: $2 input and $10 output per million tokens until November 30
gemini-4-argon is GA. Intro price $2 in, $10 out per 1M tokens through Nov 30, 2026, then $4 and $20. Sampling params are ignored.
Your keys
0 of 1 ready
Create each key on its own page, paste it below, then copy the finished lines into your project's .env file. What you paste stays in this tab. It is never sent to Wirestack or saved.
- Google Gemini API API keyGEMINI_API_KEYGet key
Click Create API key
What you get
Google's changelog says Gemini 4 Argon (gemini-4-argon) became generally available on October 13, 2026. The model page lists a 1,048,576-token context window shared by input, thinking and output. The default max_output_tokens is 65,536. Thinking levels are low, medium (the default) and high.
Introductory prices per million tokens, through November 30, 2026: $2.00 input, $0.10 cached input, $10.00 output including thinking. From December 1, 2026: $4.00 input, $0.20 cached input, $20.00 output. The model is not on the Free tier.
Breaking changes: explicit caching (cachedContents) is not supported, and temperature, top_p and top_k are ignored.
Prerequisites
Steps
- Change the model name to
gemini-4-argon. - Remove
cached_contentand any sampling parameters from your requests. - If you set
max_output_tokens, raise it; Google recommends at least 32,768. - Set client timeouts to at least 15 minutes, as the page advises for hard prompts.
Test it
Send one prompt at the low thinking level and confirm the response comes back from gemini-4-argon. Then compare medium and high on the same prompt.
Not confirmed
- Which older model, if any, Gemini 4 Argon replaces.
- Whether introductory pricing will be extended past November 30, 2026.
- Rate limits for each paid tier.