Gemini 4 Argon
Last verified October 20, 2026 · Google pricing ↗$2.00/1M input · $10.00/1M output · 1.0M context · Google
Count Tokens
—
Tokens
—
Input Cost
—
Output Cost
Estimate Monthly Cost
Monthly Cost Estimator
Quick:
<$0.0001/mo
Pick a preset above or enter custom usage
Alternatives to Gemini 4 Argon
Pricing Details
GoogleGA 2026-10-13 (Gemini API changelog), model ID gemini-4-argon. Pricing page (read 2026-10-20) now dates the switch: $2.00 input / $0.10 cache hit / $10.00 output THROUGH NOVEMBER 30, 2026; $4.00 / $0.20 / $20.00 STARTING DECEMBER 1, 2026 (page never uses the word introductory). Batch and Flex $1.00 / $0.05 / $5.00, then $2.00 / $0.10 / $10.00. Priority (1.8x) $3.60 / $0.18 / $18.00, then $7.20 / $0.36 / $36.00. Contributor tier (invite-only) = Batch. NO long-context tier: one rate to 1M (Gemini 3.1 Pro steps to $4/$18 over 200K; GPT-6.1 Sol to $4/$15 over 272K). IMPLICIT CACHING ONLY, explicit caching not supported, so no storage fee and no TTL. Not on the free tier. Search grounding 5,000 free/month shared across Gemini 3+, then $14 per 1,000. The points below about a missing end date, missing model ID, missing batch rate and partner-only access are SUPERSEDED. ORIGINAL NOTE: Announced September 30, 2026, first Gemini 4 model. INTRODUCTORY card $2.00 input, $0.10 cached input (95% off), $10.00 output, identical to GPT-6.1 Sol on all three lines. 'After the introductory period expires' it becomes $4.00/$20.00 with the same 95% cache discount ($0.20), identical to Claude Opus 5.5 on all three lines. NO end date published for the intro price (Gemini 3.8 Flash's intro card printed Dec 31), so priceSchedule is not set; budget at $4/$20. Not yet published: model ID, batch/flex rates, cache storage rate, any long-context price step (Gemini 3.1 Pro bills $4/$18 over 200K), free tier, thinking levels. Rates come from Google's launch post; as of 2026-10-01 Argon has no row on ai.google.dev pricing/models or Vertex AI pricing. ACCESS: Fairwind Program (vetted cyber defenders) and early testers only; paid Gemini API customers and AI Ultra next, no dates. Context 1M per Artificial Analysis and Vals (Google does not state it). Max output: Google says up to 1M, up from 64K, via a Long Decode Continuation API feature that resumes long answers across calls; Vals lists 262K, likely a per-call cap. Artificial Analysis Intelligence Index v4.3.2 (high, the only setting tested, read 2026-10-01): 53 at $1.99 per task at intro prices, $3.98 at $4/$20; ~110M output tokens on the index (~62K/task vs ~27K for GPT-6 Astra max). Same score as Astra max ($3.26); GPT-6.1 Sol max 52 at $0.72; Opus 5.5 high 54 at $1.82. AA-Omniscience hallucination rate 15%. Vals Index 68.90% (#1 of 41) at $15.68 per test (Vals lists $4/$20; which rate the cost uses is unstated). Google-reported (highest thinking): DeepSWE v1.1 77.9, FrontierSWE v2 55.0, Terminal-Bench 4.0 57.4, Terminal-Bench Science 0.1 57.6, AutomationBench 51.3, Harvey Legal Agent 19.6, OSWorld-2.0 offline 69.2, GraphWalks 256K-1M 84.2, LVBench 91.7.
Input / 1M tokens
$2
Output / 1M tokens
$10
Context Window
1.0M
Max Output
1.0M
Introductory pricing
Scheduled price change
Gemini 4 Argon bills at $2.00 input / $10.00 output per 1M tokens through November 30, 2026. From December 1, 2026 the rate becomes $4.00 input / $20.00 output per 1M tokens, a 100% increase. Every figure on this page uses the rate in effect today.
Google’s published pricing ↗Price History
Launched at current price on 2026-09-30. No price changes recorded since.
Frequently Asked Questions
Common questions about Gemini 4 Argon pricing and usage
Read More About Gemini 4 Argon
Kimi K3.1 Pro joins Sonnet 5.5, GPT-6.1 Sol and Gemini 4 Argon at $2 in and $10 out. Its cache hits cost double, and it spends 2.2x Sol's tokens per task, so the matching price tag hides a 3x spread.
October 24, 2026 · 7 min read
Gemini 4 Argon is out of preview, and its $2 price now has an end date: November 30. Until then it is the cheapest frontier model for prompts over 272K tokens. On December 1 it costs what Claude Opus 5.5 costs.
October 20, 2026 · 6 min read
Gemini 4 Argon launches on GPT-6.1 Sol's rate card and is scheduled to move to Claude Opus 5.5's. On Artificial Analysis, Opus 5.5 at high effort already scores higher than Argon for less than Argon's introductory price.
October 1, 2026 · 7 min read