Coding · pricing
Best for developers prototyping Gemini models with no upfront cost, and flexible usage-based pricing for production applications.
Our own captures of each vendor's public pages. Click a frame for the full-size image.
Google AI Studio has no monthly subscription price. The Free tier costs nothing, and the paid tiers are usage-based, billed per 1M tokens through the Gemini API. There are four paid channels: Standard, Batch, Flex and Priority. Enterprise is custom, sold through sales on an annual commitment. For Gemini 3.8 Flash, Standard is $0.75 input and $3.75 output per 1M tokens, Batch is $0.375 input and $1.875 output, Flex is $0.375 input and $1.875 output, and Priority is $1.35 input and $6.75 output. These rates run through Dec 31, and pricing increases on Jan 1 after that.
The Free tier includes free input and output tokens, the browser playground, the prompt gallery and a 1 million token context window. It has limited model access (experimental models like Veo are excluded) and lower rate limits than the paid tier. Free tier content is used to improve Google products. The lowest paid tier is Batch, which gives a 50% cost reduction for asynchronous, non-real-time work. Paid tier content is not used to improve Google products.
| Plan | Price | Billing | Who it is for | Key limits | API |
|---|---|---|---|---|---|
| Free Standard |
Free | Monthly | Developers learning and prototyping | Limited access to certain models (excludes experimental models like Veo); Free input and output tokens; Lower rate limits than Paid tier | Included |
| Paid Standard Standard |
Usage-based | Monthly | Production applications with standard API access | Higher rate limits for production deployments; Access to all Gemini models including experimental models; Context caching available | Included |
| Paid Batch Batch |
Usage-based | Monthly | Cost-sensitive applications with non-real-time processing | Asynchronous batch processing only; 50% cost reduction vs Standard pricing; Suitable for non-real-time workloads | Included |
| Paid Flex Flex |
Usage-based | Monthly | Applications with variable API usage patterns | Flexible rate limiting model; Variable rate limits suitable for fluctuating workloads; Lower priority than Priority tier | Included |
| Paid Priority Priority |
Usage-based | Monthly | High-traffic production applications requiring priority | Highest rate limits available; Priority processing queue; Premium support included | Included |
| Enterprise Enterprise |
Custom pricing | Monthly annual commitment |
Large enterprises with specialized requirements | Custom rate limits via provisioned throughput; Volume-based discounts available; All Paid tier features included | Included |
| If you | Pick | Why |
|---|---|---|
| you are learning or prototyping with Gemini | Free ($0) | Free input and output tokens, the playground, the prompt gallery and a 1 million token context window. Model access and rate limits are lower, and your content is used to improve Google products. |
| you run a production app with standard API access | Paid Standard (usage-based) | Access to all models, higher rate limits and context caching. Gemini 3.8 Flash costs $0.75 input and $3.75 output per 1M tokens, and your content is not used to improve Google products. |
| your workload does not need real-time answers | Paid Batch (usage-based) | Asynchronous processing at a 50% cost reduction versus Standard: $0.375 input and $1.875 output per 1M tokens for Gemini 3.8 Flash. |
| your API traffic rises and falls | Paid Flex (usage-based) | A flexible rate limiting model for variable workloads, at $0.375 input and $1.875 output per 1M tokens for Gemini 3.8 Flash. It has lower priority than the Priority tier. |
| you serve high-traffic production users | Paid Priority (usage-based) | Highest rate limits, a priority processing queue and premium support, at $1.35 input and $6.75 output per 1M tokens for Gemini 3.8 Flash. |
| you need custom throughput, security or compliance | Enterprise (contact sales) | Provisioned throughput, volume-based discounts, dedicated support channels and advanced security options, on an annual commitment. |
Free tier includes free input and output tokens, limited model access, access to Google AI Studio, and 1 million token context window; content used to improve Google products
Grounding with Google Search: 5,000 free requests per month (shared across Gemini 3.x models), then $14 per 1,000 requests. Grounding with Google Maps: 5,000 prompts per month free (shared across Gemini 3), then $14 per 1,000 queries.
Prices checked Oct 1, 2026 on the official US pricing page. Official pricing ↗ Prices and limits change often.
Vendor pricing page re-checked automatically 12 days ago: no changes detected.
Free tier includes free input and output tokens, limited model access, access to Google AI Studio, and 1 million token context window; content used to improve Google products
Free source (opens in a new tab)
Month
For: Developers learning and prototyping
Limits
Includes
API included
Usage-based source (opens in a new tab)
Month
Gemini 3.8 Flash: $0.75/1M input tokens (through Dec 31, 2026), $3.75/1M output tokens (through Dec 31, 2026); pricing increases Jan 1, 2027
For: Production applications with standard API access
Limits
Includes
API included
Usage-based source (opens in a new tab)
Month
Gemini 3.8 Flash: $0.375/1M input tokens (50% discount through Dec 31, 2026), $1.875/1M output tokens (50% discount through Dec 31, 2026)
For: Cost-sensitive applications with non-real-time processing
Limits
Includes
API included
Usage-based source (opens in a new tab)
Month
Gemini 3.8 Flash: $0.375/1M input tokens (through Dec 31, 2026), $1.875/1M output tokens (through Dec 31, 2026)
For: Applications with variable API usage patterns
Limits
Includes
API included
Usage-based source (opens in a new tab)
Month
Gemini 3.8 Flash: $1.35/1M input tokens (through Dec 31, 2026), $6.75/1M output tokens (through Dec 31, 2026); free input tokens for free tier
For: High-traffic production applications requiring priority
Limits
Includes
API included
Custom pricing source (opens in a new tab)
Month, annual commitment
Contact Sales for custom pricing
For: Large enterprises with specialized requirements
Limits
Includes
API included
Grounding with Google Search: 5,000 free requests per month (shared across Gemini 3.x models), then $14 per 1,000 requests. Grounding with Google Maps: 5,000 prompts per month free (shared across Gemini 3), then $14 per 1,000 queries.
Yes, Google AI Studio offers a free tier with limited model access and free input/output tokens. For higher rate limits, access to advanced models, and content privacy, upgrade to a Paid tier. source (opens in a new tab)
Paid tier uses usage-based pricing per 1 million tokens. For example, Gemini 3.8 Flash costs $0.75 per 1M input tokens and $3.75 per 1M output tokens through Dec 31, 2026; prices increase starting Jan 1, 2027. source (opens in a new tab)
Paid tier includes higher rate limits, access to context caching, Batch API (50% cost reduction), access to all Gemini models including experimental ones, and ensures content is not used to improve Google products. source (opens in a new tab)
Commercial use is allowed on the free tier, but Google may use your content to improve their products. For full confidentiality and no product improvement use, upgrade to a Paid tier. source (opens in a new tab)
Yes, Google offers an Enterprise tier with all Paid features plus dedicated support, advanced security & compliance, provisioned throughput, volume-based discounts, and ML ops capabilities. Contact Sales for custom pricing. source (opens in a new tab)
Yes, the Free tier has free input and output tokens, with limited model access and lower rate limits than the paid tier. Free tier content is used to improve Google products.
On Standard it is $0.75 per 1M input tokens and $3.75 per 1M output tokens. Batch and Flex are $0.375 input and $1.875 output, and Priority is $1.35 input and $6.75 output. These rates run through Dec 31, and pricing increases on Jan 1 after that.
Batch, at a 50% cost reduction versus Standard. It processes requests asynchronously, so it suits workloads that do not need real-time answers.
compare>