Gemini 4 Argon is free on Experiential Labs
Google's Gemini 4 Argon is listed as free for eligible accounts on Experiential Labs. Specs, pricing and how to test it in your AI workflows.
Experiential Labs lists Gemini 4 Argon (model id gemini-4-argon, by Google) with a Free promotion label, so you can try it in your own AI workflows without paying, as long as your account is eligible.
Key specs (from the model page)
- Context window: 1.05M tokens
- Max output: 33K tokens
- Speed: about 108 tokens/second
- Released: September 2026
- Inputs: text, image, video, PDF and audio
- List price: $2 per million input tokens, $0.10 cached, $10 per million output tokens. The promotion makes it free for eligible accounts.
What "Free" means here
The page says: "Free allowance available to eligible accounts. Check eligibility before use." So it is a free allowance, not unlimited free use for everyone. Check your account before you build on it.
How to try it
- Open the model page: https://platform.experientiallabs.ai/models/gemini-4-argon
- Click Try in Playground for a quick test, or create an API key in Settings.
- The gateway speaks the OpenAI Chat Completions API, so in most code you only change the base URL to
https://api.experientiallabs.ai/v1, use your key from theEXPLABS_API_KEYenvironment variable, and set the model togemini-4-argon.
The page also offers ready-made Python, JavaScript, cURL and "Agent prompt" quickstarts.
Good ways to test it
- Run a long-document or PDF workflow and see how the 1M context holds up.
- Try image, audio or video inputs in the same pipeline.
- Compare speed and answer quality against the model you use now before switching anything.
Because the promotion may change or end, do not make a production workflow depend on it without a fallback.