This is the closest flagship pair on price this month: identical input rates, with Grok cheaper on output and Argon cheaper on cached tokens. Argon's intro rate ends November 30, which changes the picture.
Right now Grok 4.7 and Gemini 4 Argon both charge $2.00 per 1M input tokens; Grok is cheaper on output, Argon on cache hits, so cache-heavy agents favour Argon and output-heavy jobs favour Grok. From December 1 Argon doubles, and Grok becomes cheaper on most workloads that fit in its 500K window.
| Grok 4.7 | Gemini 4 Argon | |
|---|---|---|
| Provider | SpaceXAI | |
| Tier | Flagship | Flagship |
| Input / 1M tokens | $2.00 | $2.00 |
| Output / 1M tokens | $6.00 | $10.00 |
| Cache hit / 1M | $0.50 | $0.10 |
| Cache write / 1M | — | — |
| Context window | 500K | 1M |
| Notes | New flagship (Sep 21, 2026); larger base model, replaces 4.6 at the same price; ≥200K tokens: $4/$12, cache $1.00 | 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 |
Both charge $2.00 per 1M input tokens. Grok 4.7 is 40% cheaper on output tokens ($6.00 vs $10.00 per 1M).
| Workload | Grok 4.7 | Gemini 4 Argon | Difference |
|---|---|---|---|
| Support chatbot 10,000 conversations × (2,000 input + 400 output tokens) | $64.00 | $80.00 | Grok 4.7 saves $16.00 |
| Document Q&A 1,000 questions × (30,000 input + 800 output tokens) | $64.80 | $68.00 | Grok 4.7 saves $3.20 |
| Coding agent session 4M input tokens (90% cache hits) + 150,000 output tokens | $3.50 | $2.66 | Gemini 4 Argon saves $0.84 |
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.
your workload is output-heavy, prompts stay under 200K tokens, and you are planning beyond November — Grok's rates are not scheduled to change.
you are running cache-heavy or 200K+ token work this month, you need prompts beyond 500K tokens, or you are already on Google Cloud.
Google's pricing page lists Argon at $2.00/$10.00 through November 30, 2026, and double from December 1. At that point Argon's input is 2× Grok's and its output more than 3× Grok's. Cache hits double too, to $0.20 — still below Grok's $0.50 — but with input and output now 2–3× Grok's, cached agent loops cost less on Grok 4.7 from December.
Grok 4.7 bills every token at double the rate ($4/$12) once a prompt reaches 200K tokens, and tops out at 500K. Google lists no long-context tier for Argon and a 1M window. For prompts in the 200K–500K range, Argon's intro rate is half Grok's surcharged rate this month; from December, Argon's $4/$20 matches Grok's surcharged input and costs more on output. Above 500K, only Argon fits.
Argon caches implicitly — repeated prefixes are billed at $0.10 with no write fee. Grok 4.7 also lists no write fee, but its cache hit ($0.50) is a quarter of input rather than a twentieth. Argon is not on the Gemini API free tier.
In November 2026 they share an input price; Grok 4.7 is cheaper on output and Gemini 4 Argon on cache hits. After Argon's price doubles on December 1, Grok 4.7 is cheaper on standard rates.
Gemini 4 Argon, with a 1M window and no long-context tier. Grok 4.7 caps at 500K tokens and doubles rates at 200K.
No. Google's pricing page says it is not available on the free tier.
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.