Worked examples
The simulator's four example sessions on Gemini 4 Argon with implicit caching, priced turn by turn by the same cost model. Best case: Google publishes no lifetime for implicit caching and does not guarantee hits, so this assumes every eligible turn hits.
-
Coding agent
Tool calls and edits in quick succession, with a few longer pauses. 40 turns, 20K tokens of starting context, 3K in and 600 out per turn, 1.5 minutes apart, three long breaks of 8 minutes.
Open this in the simulator- With caching
- $0.92
- Without
- $7.70
- Saved
- $6.77 88%
- Pays for itself on turn 2
-
Long chat
A person reading and typing, with one long break. 30 turns, 3K tokens of starting context, 150 in and 500 out per turn, 3 minutes apart, one long break of 45 minutes.
Open this in the simulator- With caching
- $0.35
- Without
- $0.90
- Saved
- $0.56 62%
- Pays for itself on turn 4
-
Big document
One large document, questioned a dozen times. 12 turns, 150K tokens of starting context, 100 in and 700 out per turn, 7 minutes apart.
Open this in the simulator- With caching
- $0.61
- Without
- $3.79
- Saved
- $3.18 84%
- Pays for itself on turn 2
-
Unattended loop
A script calling the model nonstop, the prompt growing every turn. 200 turns, 30K tokens of starting context, 1.5K in and 300 out per turn, 30 seconds apart.
Open this in the simulator- With caching
- $6.33
- Without
- $84.84
- Saved
- $78.51 93%
- Pays for itself on turn 2
When caching pays for itself
For a prefix sent again and again inside the cache lifetime. The count does not depend on the prefix size, as long as it is above the minimum.
Implicit (automatic)
There is no write charge. A read costs $0.10 per million tokens instead of $2.00, so caching never costs more than sending the prefix uncached.
A cached prefix pays for itself after 1 read.
Best case Google publishes no lifetime and no guarantee of a hit, so this shows the best case: every eligible turn hits.
Caching rules for Gemini 4 Argon
- Implicit caching is on by default for Gemini 2.5 and newer, with no write charge, no storage charge and no cost-saving guarantee. A hit costs 0.05x the input price. Google publishes no lifetime, so implicit figures here are a best case in which every eligible turn hits.
- The implicit caching minimum is 4,096 tokens.
- Gemini 4 caches the full previous request only if the new request adds at least 2,050 tokens. Otherwise the last 2,050 tokens of the previous request are reprocessed at the standard input price.
- Explicit caching is not available on Gemini 4 Argon, only on Gemini 3 and 2.5 models.
- The cached price is 0.05x the input price on the Gemini API pricing page, while Google Cloud's overview describes a bigger discount. Cachemeter follows the pricing page.
- Google's pricing page lists a rise on 1 December 2026, to $4.00 input, $0.20 cached and $20.00 output. The simulator switches on that date.
Sources: Google's pricing page, caching guide. See all prices and sources.
Similar models
Per million tokens, with the 40-turn coding agent on each model's default cache setting.
| Model | Input | Cache read | Output | Coding agent |
|---|---|---|---|---|
| Gemini 4 Argon | $2.00 | $0.10 | $10.00 | $0.92 best case* |
| Gemini 3.1 Pro Preview | $2.00 | $0.20 | $12.00 | $1.33 best case* |
| Gemini 2.5 Pro | $1.25 | $0.125 | $10.00 | $0.89 best case* |
| Claude Sonnet 5.5 | $2.00 | $0.10 | $10.00 | $1.77 |
| GPT-6.1 Sol | $2.00 | $0.10 | $10.00 | $1.00 |
* Gemini implicit caching, best case.
Questions
How much does context caching cost on Gemini 4 Argon?
Input costs $2.00 per million tokens, with no charge to write the cache, $0.10 to read it and $10.00 for output.
Is prompt caching worth it for Gemini 4 Argon?
In the 40-turn coding agent example, caching costs $0.92 against $7.70 without, a saving of $6.77 (88%). With implicit caching there is no write charge, so caching never costs more than not caching. This is a best case, because Google does not guarantee hits.
How long does the Gemini 4 Argon cache last?
Implicit (automatic): Google publishes no lifetime and no guarantee of a hit, so this shows the best case: every eligible turn hits.
Does Gemini 4 Argon have a minimum cacheable prompt length?
Yes. Prompts shorter than 4,096 tokens are not cached.
An estimate from published list prices, and your bill may differ. Token counts are what the example sets, not tokenized text. Not covered: batch and fast-mode prices, thinking tokens beyond the output shown, tool fees and regional uplifts.