Gemini 4 Argon became generally available on Oct 13, 2026 at an introductory price that undercuts Gemini 3.1 Pro on output. That introductory rate ends on November 30, 2026, after which Argon doubles.
Through November 30, Gemini 4 Argon is the cheaper model: same $2.00 input, lower output ($10.00 vs $12.00) and half the cache-hit price. From December 1 Argon moves to $4/$20 per 1M, making Gemini 3.1 Pro cheaper for prompts up to 200K tokens.
| Gemini 4 Argon | Gemini 3.1 Pro | |
|---|---|---|
| Provider | ||
| Tier | Flagship | Flagship |
| Input / 1M tokens | $2.00 | $2.00 |
| Output / 1M tokens | $10.00 | $12.00 |
| Cache hit / 1M | $0.10 | $0.20 |
| Cache write / 1M | — | — |
| Context window | 1M | 1M |
| Notes | New top frontier model, generally available Oct 13, 2026 for AI Pro/Ultra subscribers and via the API (model id gemini-4-argon); long-decode generation up to 1M tokens; intro price thru Nov 30, 2026, then $4/$20 (cache read $0.20); not available on the free tier | Frontier multimodal (>200K: $4/$18, cache $0.40) |
Both charge $2.00 per 1M input tokens. Gemini 4 Argon is 17% cheaper on output tokens ($10.00 vs $12.00 per 1M).
| Workload | Gemini 4 Argon | Gemini 3.1 Pro | Difference |
|---|---|---|---|
| Support chatbot 10,000 conversations × (2,000 input + 400 output tokens) | $80.00 | $88.00 | Gemini 4 Argon saves $8.00 |
| Document Q&A 1,000 questions × (30,000 input + 800 output tokens) | $68.00 | $69.60 | Gemini 4 Argon saves $1.60 |
| Coding agent session 4M input tokens (90% cache hits) + 150,000 output tokens | $2.66 | $3.32 | Gemini 4 Argon saves $0.66 |
Standard rates per 1M tokens from data on this site. The coding-agent row assumes 90% of input is read from the prompt cache (full input price where a model has no cache rate). Cache-write fees, batch discounts and long-context surcharges listed in the notes are not included. Updated November 2, 2026.
you want Google's newest frontier model, need long-decode generation of up to 1M tokens, or are running large jobs before the intro price ends on Nov 30.
you are budgeting for December onward and most prompts are under 200K tokens, where 3.1 Pro's $2.00/$12.00 will undercut Argon's post-intro rate.
Until Nov 30, 2026 Argon costs $2.00 input and $10.00 output with $0.10 cache hits. From Dec 1 it becomes $4/$20 with $0.20 cache hits. Gemini 3.1 Pro stays at $2.00/$12.00 for prompts up to 200K tokens and $4/$18 above. After the switch, 3.1 Pro is cheaper at every prompt length on output and at short lengths on input.
Google lists no long-context tier for Argon, while 3.1 Pro steps up past 200K tokens. Argon's caching is implicit only, with no hourly storage fee listed; 3.1 Pro supports explicit context caching, which carries an hourly storage charge. Neither model is on the free tier.
Google's pricing page still labels 3.1 Pro as a Preview model, while Argon is generally available. If you need a GA model for production commitments, that may matter as much as price. Both get Google's 50% batch discount.
Until November 30, 2026, yes on output and cache hits, with equal input. From December 1, Argon rises to $4/$20 per 1M and 3.1 Pro becomes cheaper for prompts under 200K tokens.
November 30, 2026. Google's pricing page shows input, output and caching rates doubling from December 1.
Yes, prompts over 200K tokens are billed at $4 input and $18 output per 1M.
No. Google lists no free tier for Argon or for Gemini 3.1 Pro.
How we choose: prices come from each vendor's official pricing page and are re-checked every week. We take no affiliate fees, so rankings aren't for sale.