This free calculator compares the API cost of Gemini 4 Argon and Claude Opus 5.5 for your own token counts, per request and per month. Argon (gemini-4-argon) costs $2 input / $10 output per million tokens until November 30, 2026, then $4 / $20 from December 1, 2026. Claude Opus 5.5 (claude-opus-5-5) costs $4 / $20. So from December the list prices match, and the difference comes down to caching, Batch and Fast mode, and how many tokens each model uses for your task. Prices come from Google's and Anthropic's pricing pages, checked October 19, 2026.
Prices used by the calculator
All prices are per 1 million tokens, from Google’s Gemini API pricing page and Anthropic’s Claude pricing page, checked on 19 October 2026.
| Gemini 4 Argon | Input | Output (incl. thinking) | Cache hit |
|---|---|---|---|
| Standard, through 30 Nov 2026 | $2.00 | $10.00 | $0.10 |
| Standard, from 1 Dec 2026 | $4.00 | $20.00 | $0.20 |
| Batch or Flex, through 30 Nov / from 1 Dec | $1.00 / $2.00 | $5.00 / $10.00 | $0.05 / $0.10 |
| Priority, through 30 Nov / from 1 Dec | $3.60 / $7.20 | $18.00 / $36.00 | $0.18 / $0.36 |
| Claude Opus 5.5 | Input | Output | Cache read | Cache write (5 min / 1 h) |
|---|---|---|---|---|
| Standard | $4.00 | $20.00 | $0.20 | $5.00 / $8.00 |
| Batch API | $2.00 | $10.00 | $0.10 | $2.50 / $4.00 |
| Fast mode | $8.00 | $40.00 | $0.40 | $10.00 / $16.00 |
How to use it
- Type the input and output tokens of one typical request, or tap a preset.
- Set the cached share if part of every prompt repeats (a long system prompt, a fixed document).
- Enter how many requests you send in a month to get a monthly bill.
- Pick the Argon tier and the Opus mode. Tick the cache-write box if each Opus request writes a new cache (for example, the first call of every session).
Things the numbers include and leave out
- Argon has no separate cache-write fee. Its caching is implicit and discounts repeated input by 95%.
- Argon output prices include thinking tokens. Opus 5.5 always uses adaptive thinking, and its thinking tokens are billed as output, so put your expected thinking in the output box for both.
- Opus 5.5 charges the same price across its full 1M context. Fast mode can’t be combined with the Batch API.
- Grounding with Google Search (5,000 free prompts a month, then $14 per 1,000) and Claude tool fees are not included.
Gemini 4 Argon vs Claude Opus 5.5 at a glance
| Gemini 4 Argon | Claude Opus 5.5 | |
|---|---|---|
| Model id | gemini-4-argon |
claude-opus-5-5 |
| Maker | Anthropic | |
| Available since | October 13, 2026 (generally available) | September 22, 2026 |
| Input / output per 1M tokens | $2 / $10 until Nov 30, 2026; $4 / $20 from Dec 1, 2026 | $4 / $20 |
| Cheaper repeat input | Implicit caching only: 95% off ($0.10, then $0.20 from Dec 1) | Cache reads $0.20; cache writes $5 (5 min) or $8 (1 hour) |
| Half-price async option | Batch and Flex, −50% | Batch API, −50% |
| Faster option | Priority: $3.60 / $18 (then $7.20 / $36) | Fast mode: $8 / $40 |
| Context window | 1M tokens | 1M tokens, same price across the whole window |
| Max output | Up to about 1M tokens (long decode) | 128K tokens (300K on the Batch API beta) |
| Free tier | None. Paid tier only | Paid API |
| Thinking | Levels low / medium / high (default medium); thinking tokens billed as output | Adaptive thinking, always on (default effort medium); billed as output |
What changes on December 1, 2026
Google lists Argon's current prices as an introductory price through November 30, 2026. From December 1, 2026, every Argon price doubles: Standard goes from $2 / $10 to $4 / $20, Batch and Flex from $1 / $5 to $2 / $10, Priority from $3.60 / $18 to $7.20 / $36, and cache hits from $0.10 to $0.20. The calculator shows both prices side by side so you can budget for December.
Which one is cheaper?
- Until November 30: Argon costs half as much as Opus 5.5 at list price for the same number of tokens.
- From December 1: their standard input and output prices are the same. Argon still wins when a lot of your prompt repeats, because its caching is automatic with no write fee. Opus 5.5 can come out ahead if you write a cache once and reuse it many times within its lifetime, or if it finishes a task in fewer tokens.
- Token counts aren't directly comparable. Each company uses its own tokenizer. Anthropic says the tokenizer used by Claude 4.7 and later models produces about 30% more tokens for the same text than its earlier one. Measure a few real requests on both before you commit.
Frequently asked questions
How much does Gemini 4 Argon cost?
$2 per million input tokens and $10 per million output tokens (thinking included) until November 30, 2026. From December 1, 2026 it is $4 input and $20 output. Batch and Flex are half price.
Is Gemini 4 Argon going up in price?
Yes. Google lists the current prices as introductory through November 30, 2026. From December 1 every Argon rate doubles, including Batch, Flex, Priority and cache hits.
How much does Claude Opus 5.5 cost?
$4 per million input tokens and $20 per million output tokens, across its full 1M-token context. Cache reads are $0.20, Batch API is half price, and Fast mode is $8 / $40.
Is Gemini 4 Argon cheaper than Claude Opus 5.5?
Until November 30, 2026, yes: half the list price. From December 1 the standard input and output prices are the same, so the cheaper one depends on how much of your prompt is cached and how many tokens each model uses.
Does Gemini 4 Argon have a free tier?
No. Google lists Argon on the paid tier only.
Is there a long-context surcharge?
Not on either model. Both have a 1M-token context, and Anthropic says Opus 5.5 is billed at the standard price across the full window.
Are thinking tokens billed?
Yes, on both. Argon's output price includes thinking tokens, and Opus 5.5's adaptive thinking is billed as output. Add your expected thinking to the output tokens in the calculator.
Does the calculator include Google Search grounding?
No. Grounding with Google Search on Argon has 5,000 free prompts a month, then costs $14 per 1,000. Add it separately if you use it.