Gemini 4 Argon Goes GA — With a Six-Week Pricing Cliff That Lands Exactly on Claude Opus 5.5's Rate

Gemini 4 Argon Goes GA — With a Six-Week Pricing Cliff That Lands Exactly on Claude Opus 5.5's Rate

Update (Oct 17): the pricing cliff is now the story. Google released Gemini 4 Argon to developers and businesses worldwide on October 13, 2026, ending the restricted "Fairwind" pre-access regime — Google's own announcement: "Today, we're announcing our new frontier model, Gemini 4 Argon, which is rolling out to a set of trusted cyber defenders through our Fairwind Program," with cyber defenders as the priority early-access category ahead of the general rollout. Google's developer documentation describes it bluntly: "Gemini 4 Argon is our latest, most capable model."

The new detail from this week's documentation deep-dive: Argon's introductory API pricing runs only through November 30, 2026 — $2/M input, $10/M output, $0.10/M cached input (a 95% cached-input discount) — and doubles on December 1 to $4/$20/$0.20. That standard rate is exactly Claude Opus 5.5's per-token price, so the six-week discount is a wedge: "a six-week discount window that expires right as a rival's pricing becomes directly comparable looks like a deliberate wedge to pull developers into the Gemini ecosystem while the discount lasts¹, then rely on switching costs once standard pricing resets."

Also confirmed in the documentation this week:

• Unified 1M-token context: input and output pool into a single 1M envelope (vs. split budgets on earlier Geminis), with per-call caps of 32K (non-streaming) and 64K (streaming) — built for long-horizon agentic sessions in Antigravity, not chat.
• Access-tier confusion: Google AI Pro subscribers can use Argon in the Gemini app and AI Mode in Search, but per Google's own API docs Argon is not available on the Google AI Pro plan for API access and is excluded from the Free tier entirely — a gap that "has tripped up more than a few developers this week."
• Adoption and skepticism: GitLab added Argon to its Duo agent platform within days of GA; Gartner is questioning whether real-world enterprise performance will justify the cost once intro pricing expires.

What it means: the price war (OpenAI Answers Opus 5.5 Same-Day with GPT-6 Sol and Luna at 50% Cuts — the Price War Moves to Cost-per-Task) has moved from launch-day discounts to timed cliffs designed to manufacture lock-in before December 1. The December check: whether enterprises stay at $4/$20 Argon or the intro window proves to be a trial funnel — and whether Anthropic/OpenAI respond with matching promotional windows.