Google Gemini API pricing for WhatsApp chatbots and voice agents
As on 17 September 2026, Google lists Gemini API prices in US dollars per 1 million tokens. Gemini 2.5 Flash-Lite is $0.10 input and $0.40 output, and Gemini 3.8 Flash is $0.75 and $3.75 until 31 December 2026. Gemini 3.8 Live audio input costs $0.005 per minute. Most models also have a rate-limited free tier. On Convo360 you connect your own Gemini key with zero markup.
Prices as on 17 September 2026. Figures were checked against the provider's official pages on that date. Providers change prices and rules without notice, so confirm on the official page before you budget. Convo360 adds no markup: you pay the provider directly with your own account. Convo360 is not affiliated with Google. Google, Gemini and other product names are trademarks of their owner. Prices were read from Google's official pages on 17 September 2026 in USD and may change; always confirm on the official pricing page.
- Google bills the Gemini API per token in USD, and output prices include the model's thinking tokens.
- Most Flash and Flash-Lite models have a free tier, but Google says free tier content is used to improve its products.
- Gemini 3.8 Flash is $0.75 input and $3.75 output per 1 million tokens until 31 December 2026, then $1.50 and $7.50.
- Gemini 3.8 Live, used for voice calls, lists audio at $0.005 per minute input and $0.018 per minute output.
- On Convo360 your Gemini key goes in Settings → Provider Keys, and Gemini Live can be the speech engine for calls.
Gemini API prices as on 17 September 2026
These figures were read from Google's official Gemini API pricing page on 17 September 2026. They are Standard, Paid Tier prices in US dollars per 1 million tokens. Output prices include thinking tokens. The table covers text models a business chatbot is likely to use.
| Model | Input | Output | Context caching | Free tier |
|---|---|---|---|---|
| Gemini 3.8 Flash, 3.7 Flash, 3.6 Flash | $0.75 | $3.75 | $0.075 | Yes |
| Gemini 3.5 Flash | $1.50 | $9.00 | $0.15 | Yes |
| Gemini 3.5 Flash-Lite | $0.30 | $2.50 | $0.03 | Yes |
| Gemini 3.1 Flash-Lite | $0.25 text, $0.50 audio | $1.50 | $0.025 | Yes |
| Gemini 3.1 Pro Preview | $2.00 | $12.00 | $0.20 | No |
| Gemini 2.5 Flash | $0.30 text, $1.00 audio | $2.50 | $0.03 | Yes |
| Gemini 2.5 Flash-Lite | $0.10 text, $0.30 audio | $0.40 | $0.01 | Yes |
Two notes from the same page. The Gemini 3.8, 3.7 and 3.6 Flash prices apply through 31 December 2026; from 1 January 2027 Google lists $1.50 input, $7.50 output and $0.15 caching. Gemini 3.1 Pro Preview is a preview model with a context-length band: prompts above 200k tokens cost $4.00 input and $18.00 output. Cached tokens also carry an hourly storage price. Convo360 is not affiliated with Google.
Live API prices for voice calls
Google describes its Live models as low-latency, audio-to-audio models for real-time voice agents. Gemini 3.8 Live, Gemini 3.8 Live Extended Thinking and Gemini 3.1 Flash Live Preview share one price table:
| Item | Per 1 million tokens | Per minute, as listed |
|---|---|---|
| Text input | $0.75 | |
| Audio input | $3.00 | $0.005 |
| Text output | $4.50 | |
| Audio output | $12.00 | $0.018 |
The older Gemini 2.5 Flash Native Audio preview model lists $0.50 text and $3.00 audio input, and $2.00 text and $12.00 audio output. All these Live models show a free tier.
ILLUSTRATIVE call: 3 minutes of caller audio is 3 × $0.005 = $0.015. If the model replies as text, 600 output tokens add about $0.0027. If Gemini also speaks for 1.5 minutes, add 1.5 × $0.018 = $0.027. Telephony minutes are billed by your telephony provider. For how this engine differs from the classic chain, see Gemini Live for voice agents.
Free tier versus paid tier, as Google publishes it
- Free tier: limited access to certain models, free input and output tokens, and Google AI Studio access. Google states that free tier content is used to improve its products, which matters if real customer chats pass through it.
- Paid tier: higher rate limits, context caching, the Batch API at a 50% cost reduction, and content not used to improve Google's products.
- How free limits work: limits are counted in requests per minute, input tokens per minute and requests per day. They apply per project, not per API key, and daily quotas reset at midnight Pacific time. Preview models are more restricted.
- The actual numbers: Google no longer prints per-model free limits in its docs. You view them inside AI Studio, so we do not quote any.
To move to the paid tier you link a billing account and prepay a minimum of $5. Google says unused prepaid credits expire after 12 months and are non-refundable. Tier 1 accounts carry a $250 monthly spend cap, rising to $2,000 at Tier 2.
A per-conversation formula with illustrative numbers
Google's guide says a Gemini token is about 4 characters, and 100 tokens is about 60 to 80 English words.
Cost per conversation = (input tokens ÷ 1,000,000 × input price) + (output tokens ÷ 1,000,000 × output price).
The counts below are ILLUSTRATIVE: a WhatsApp chat of about eight messages using 6,000 input tokens in total and 800 output tokens, thinking tokens included. Use your own figures from AI Studio.
| Model | Working | One chat | 1,000 chats |
|---|---|---|---|
| Gemini 2.5 Flash-Lite | $0.0006 input + $0.00032 output | $0.00092 | $0.92 |
| Gemini 3.5 Flash-Lite | $0.0018 input + $0.002 output | $0.0038 | $3.80 |
| Gemini 3.8 Flash, to 31 Dec 2026 | $0.0045 input + $0.003 output | $0.0075 | $7.50 |
The full method for measuring tokens per conversation is in estimating your AI provider bill.
Paying Google in USD from India
The Gemini price list is in US dollars, and Google's billing page describes the $5 minimum prepayment as $5 or the equivalent in other currencies. We do not convert prices, because the rate that matters is the one applied on the day you pay:
₹ cost = USD amount × your bank's rate on the day + GST and forex fees, if any.
- Neither the pricing page nor the billing page says whether taxes are included or extra. Check your first invoice before fixing a budget.
- If you pay by card, it must allow international transactions. Forex mark-up varies by bank, so ask yours.
- Google Cloud's $300 welcome credit cannot be used for the Gemini API, according to the billing FAQ.
Adding your Gemini key in Convo360, for chat and for calls
Convo360 is BYOK: Google bills your own account and Convo360 adds zero markup on tokens.
- Create an API key in Google AI Studio. Set up billing there if you want the paid tier.
- In Convo360, open Settings → Provider Keys, paste the key under Google Gemini and click Test connection. Keys are encrypted at rest.
- Open your AI agent and choose a Gemini model in AI model (text replies). The model is set per agent.
- For phone calls, set Speech engine (hears the caller) to Gemini Live. It hears the caller and thinks, while the voice you picked, such as Sarvam, Smallest.ai or ElevenLabs, speaks. That voice provider bills its part separately. Voice calls need the Pro plan at ₹2,999 per month, excluding 18% GST.
- Test with Preview / Try-it, then place a real call.
Match the model name shown on your agent to the row on Google's page, because the list changes as Google adds and retires models. When prepaid credit reaches $0, Google says every API key on that billing account stops working. Convo360's Provider Health monitoring detects the errors and shows a clear alert. See the AI agents docs and how a BYOK platform works.
Ways to cut the Gemini bill
- Watch thinking tokens. They are billed as output, the dearer side of every row.
- Keep the opening of your prompt stable. Google says implicit caching is on by default for Gemini 2.5 and newer models, and cached tokens are listed far below the input price.
- Send fewer knowledge base passages and cap chat history.
- Set a project spend cap in AI Studio. Google marks it experimental, with roughly ten minutes of billing delay.
- Use Batch for offline work. Half price, but not meant for live replies.
- Avoid Google Search grounding unless needed. On Gemini 3 models it is $14 per 1,000 requests after 5,000 free requests a month.
- Use guided flows for menus and forms. They use no tokens.
Comparing providers on more than price? Read OpenAI vs Claude vs Gemini for chatbots.
Example conversation
Connect your own Gemini key, zero markup
Start free in minutes — no code, no credit card. Bring your own AI keys and pay no per-message markup.
Frequently asked questions
Is there a free tier on the Gemini API?
Yes. As on 17 September 2026, Google lists a free tier with free input and output tokens on most Flash and Flash-Lite models and on the Live models, within rate limits shown in AI Studio. Gemini 3.1 Pro Preview has no free tier. Free tier content is used to improve Google's products.
What are the Gemini free tier rate limits?
Google measures them in requests per minute, input tokens per minute and requests per day, per project, with daily quotas resetting at midnight Pacific time. The per-model numbers are no longer printed in the docs. You view your live limits inside Google AI Studio.
Is GST extra on Gemini API prices in India?
Google's Gemini pricing and billing pages, as read on 17 September 2026, do not say whether taxes are included or extra. Prices are published in US dollars. Check the first invoice on your billing account before you finalise a rupee budget.
What does Gemini Live cost per minute for a voice call?
For Gemini 3.8 Live, Google lists audio input at $0.005 per minute and audio output at $0.018 per minute, plus $0.75 input and $4.50 output per 1 million text tokens. Telephony minutes and any separate voice provider are billed on top by those providers.
Will Gemini 3.8 Flash prices change in 2027?
Yes, according to Google's own page. Gemini 3.8, 3.7 and 3.6 Flash are listed at $0.75 input and $3.75 output per 1 million tokens through 31 December 2026, and at $1.50 input and $7.50 output starting 1 January 2027.
Can I use my own Gemini key with Convo360?
Yes. Convo360 is BYOK, so you paste your own Gemini API key under Settings → Provider Keys, choose a Gemini model on each agent, and optionally set Gemini Live as the speech engine for calls. Google bills you directly and Convo360 adds zero markup.
Official sources
- Gemini API docs: Pricing (official)
- Gemini API docs: Rate limits and usage tiers (official)
- Gemini API docs: Billing, prepay and spend caps (official)
- Gemini API docs: Understand and count tokens (official)
- Gemini API docs: Context caching (official)
Related reading
Available across India — see WhatsApp & AI calling automation by state and city.