This Gemini API cost calculator estimates paid-tier text-token charges for Gemini 4 Argon and Gemini Flash models. Choose Standard, Batch, Flex or Priority processing, separate uncached and cached input, then multiply a representative request by your expected volume.
Google publishes Standard, Batch, Flex and Priority text-token rates. This planner does not include storage, grounding or media-specific pricing. Confirm production budgets against the official Gemini Developer API pricing page.
Estimate your Gemini API cost
Pricing and model routing verified: 19 October 2026. Paid-tier rates in USD.
Gemini 3.8 Flash · Standard · 1,000 requests
- Uncached input
- $0.75
- Cached input
- $0.00
- Output
- $1.875
- Average per request
- $0.002625
Planning estimate only. It excludes cache storage time, grounding, media-specific differences, retries, taxes and contracted discounts.
Current Gemini API rates used by this calculator
All rates are paid-tier USD per million tokens for text tokens. The calculator includes Standard, Batch, Flex and Priority processing.
| Model | Mode | Input | Cached input | Output |
|---|---|---|---|---|
| Gemini 4 Argon | Standard | $2.00 | $0.10 | $10.00 |
| Gemini 4 Argon | Batch | $1.00 | $0.05 | $5.00 |
| Gemini 4 Argon | Flex | $1.00 | $0.05 | $5.00 |
| Gemini 4 Argon | Priority | $3.60 | $0.18 | $18.00 |
| Gemini 3.8 Flash | Standard | $0.75 | $0.075 | $3.75 |
| Gemini 3.8 Flash | Batch | $0.375 | $0.0375 | $1.875 |
| Gemini 3.7 Flash (routes to 3.8) | Standard | $0.75 | $0.075 | $3.75 |
| Gemini 3.7 Flash (routes to 3.8) | Batch | $0.375 | $0.0375 | $1.875 |
| Gemini 3.6 Flash | Standard | $0.75 | $0.075 | $3.75 |
| Gemini 3.6 Flash | Batch | $0.375 | $0.0375 | $1.875 |
| Gemini 3.5 Flash (routes to 3.6) | Standard | $0.75 | $0.075 | $3.75 |
| Gemini 3.5 Flash (routes to 3.6) | Batch | $0.375 | $0.0375 | $1.875 |
| Gemini 3.5 Flash-Lite | Standard | $0.30 | $0.03 | $2.50 |
| Gemini 3.5 Flash-Lite | Batch | $0.15 | $0.02 | $1.25 |
| Gemini 3.1 Flash-Lite | Standard | $0.25 | $0.025 | $1.50 |
| Gemini 3.1 Flash-Lite | Batch | $0.125 | $0.0125 | $0.75 |
| Gemini 3.8 Flash | Flex | $0.375 | $0.0375 | $1.875 |
| Gemini 3.8 Flash | Priority | $1.35 | $0.135 | $6.75 |
| Gemini 3.7 Flash (routes to 3.8) | Flex | $0.375 | $0.0375 | $1.875 |
| Gemini 3.7 Flash (routes to 3.8) | Priority | $1.35 | $0.135 | $6.75 |
| Gemini 3.6 Flash | Flex | $0.375 | $0.0375 | $1.875 |
| Gemini 3.6 Flash | Priority | $1.35 | $0.135 | $6.75 |
| Gemini 3.5 Flash (routes to 3.6) | Flex | $0.375 | $0.0375 | $1.875 |
| Gemini 3.5 Flash (routes to 3.6) | Priority | $1.35 | $0.135 | $6.75 |
| Gemini 3.5 Flash-Lite | Flex | $0.15 | $0.02 | $1.25 |
| Gemini 3.5 Flash-Lite | Priority | $0.54 | $0.05 | $4.50 |
| Gemini 3.1 Flash-Lite | Flex | $0.125 | $0.0125 | $0.75 |
| Gemini 3.1 Flash-Lite | Priority | $0.45 | $0.045 | $2.70 |
Time-limited rates: Google lists the Gemini 3.8 and 3.6 Flash prices above through 31 December 2026, including the routed legacy IDs shown here. Those token rates double on 1 January 2027. The calculator does not silently project the future prices.
Gemini 4 Argon: The listed promotional token rates apply through 30 November 2026 and double on 1 December 2026. It has no free tier and supports implicit cache hits only, not explicit cache creation or storage. The shared input/output limit is 1,048,576 tokens; output above the default synchronous limit of 65,536 requires long decode. This estimates token charges only, not private Contributor-tier or tool fees. See the official Argon model limits. Recheck prices before budgeting beyond the promotion.
Deprecated model IDs: Google announced on 8 October 2026 that API requests to gemini-3.7-flash route to gemini-3.8-flash, and requests to gemini-3.5-flash route to gemini-3.6-flash. The endpoints are not shutting down. These options use the destination model’s current published rates, not the old model’s historical prices. For new work, select the replacement ID. See Google’s 8 October release notes and deprecation schedule.
How the Gemini estimate works
- Choose the model and mode. Batch is for asynchronous workloads. Flex and Priority use their own published token rate schedules.
- Enter total input and the cached portion. The uncached amount is total input minus cached input.
- Enter output tokens. Google’s published output price includes thinking tokens for these models.
- Multiply by request volume. Use a typical daily or monthly volume, then reconcile the estimate against billed usage.
Formula used
Total = requests × (((input - cached input) × input rate) + (cached input × cached rate) + (output × output rate)) / 1,000,000
What this estimate does not include
- Context-cache storage charged per million tokens per hour.
- Grounding with Google Search or Google Maps after any included allowance.
- Managed-agent environment fees and enterprise platform prices.
- Modality-specific differences such as Gemini 3.1 Flash-Lite audio input rates.
- Taxes, currency conversion, retries and negotiated pricing.
Frequently asked questions
Is the Gemini API free?
Some models offer a free tier subject to Google’s current limits and data-use terms. This calculator estimates the paid tier only.
Does this calculator support images or audio?
It is intentionally a text-token planner. Some models price other modalities at the same token rate, while others differ. Use Google’s modality-specific table for mixed-media production estimates.
Why can the final bill differ?
Thinking tokens, grounding, cache storage, retries, tool use, processing mode and future rate changes can all alter the final charge.
Compare providers with the Claude API cost calculator and OpenAI API cost calculator, or visit the AI cost calculators hub.