AI API pricing in the Philippines: Gemini, OpenAI and Claude in pesos
Short answer
Gemini, OpenAI and Claude all bill their APIs per 1 million tokens in US dollars, and all three list the Philippines as a supported country. On 12 Oct 2026 the low-cost models start near US$0.10 per million input tokens (about PHP 6), and the mid-range models sit near US$2 (about PHP 126). Google offers a free tier on most Gemini models. Prices change often, so confirm on each vendor’s page.
Prices at a glance
Peso amounts are approximate and use the rate shown at the top of this page.
Gemini 2.5 Flash-Lite
PHP 6 / PHP 25 Approx. pesos in / out- Input per 1M tokens US$0.10
- Output per 1M tokens US$0.40
Gemini 2.5 Flash
PHP 19 / PHP 157 Approx. pesos in / out- Input per 1M tokens US$0.30
- Output per 1M tokens US$2.50
Gemini 3.1 Pro Preview (prompts up to 200k tokens)
PHP 126 / PHP 754 Approx. pesos in / out- Input per 1M tokens US$2.00
- Output per 1M tokens US$12.00
OpenAI gpt-6-luna
PHP 6 / PHP 31 Approx. pesos in / out- Input per 1M tokens US$0.10
- Output per 1M tokens US$0.50
OpenAI gpt-6.1-sol
PHP 126 / PHP 628 Approx. pesos in / out- Input per 1M tokens US$2.00
- Output per 1M tokens US$10.00
OpenAI gpt-6-astra
PHP 628 / PHP 3,141 Approx. pesos in / out- Input per 1M tokens US$10.00
- Output per 1M tokens US$50.00
Showing the first 6 of 9 rows. The full table with features and limits follows below.
On this page
AI API pricing basis
- Billing unit
- US dollars per 1 million tokens, input and output priced separately
- Currency named by the vendor
- Google and Anthropic state USD. OpenAI shows a “$” sign and does not name a code
- Exchange rate used
- USD 1 = PHP 62.821, from the ECB euro reference rates for 9 Oct 2026
- Philippines supported
- Yes on all three vendors’ country lists
- Batch discount
- 50 percent on all three, per each vendor’s page
- Last verified
- 12 Oct 2026
What the API is, and how it differs from the chat apps
The chat apps are products a person opens: Gemini, ChatGPT and Claude. The API is a separate product for developers and workflow tools. A person signs in, creates a key, and software uses the key to call a model. Each vendor’s pricing page lists the API as its own per-token product, apart from the chat plans. Check the billing terms before assuming a chat subscription covers API calls.
The n8n and Zapier pages cover the tools that most often make these calls. Those tools charge for the workflow. The model provider charges for the model.
Prices per 1 million tokens
A token is a small piece of text, often a short word or part of one. The vendors price input and output separately. The figures below are standard rates for text, as listed on each vendor’s pricing page on 12 Oct 2026. Peso amounts are conversions, labelled approximate.
| Vendor and model | Input per 1M tokens | Output per 1M tokens | Approx. pesos in / out |
|---|---|---|---|
| Gemini 2.5 Flash-Lite | US$0.10 | US$0.40 | PHP 6 / PHP 25 |
| Gemini 2.5 Flash | US$0.30 | US$2.50 | PHP 19 / PHP 157 |
| Gemini 3.1 Pro Preview (prompts up to 200k tokens) | US$2.00 | US$12.00 | PHP 126 / PHP 754 |
| OpenAI gpt-6-luna | US$0.10 | US$0.50 | PHP 6 / PHP 31 |
| OpenAI gpt-6.1-sol | US$2.00 | US$10.00 | PHP 126 / PHP 628 |
| OpenAI gpt-6-astra | US$10.00 | US$50.00 | PHP 628 / PHP 3,141 |
| Claude Haiku 4.5 | US$1 | US$5 | PHP 63 / PHP 314 |
| Claude Sonnet 5.5 | US$2 | US$10 | PHP 126 / PHP 628 |
| Claude Opus 5.5 | US$4 | US$20 | PHP 251 / PHP 1,256 |
Swipe the table sideways to see all columns.
These are a sample, not a full list. Each page also lists older models, audio, image and video models, caching rates and tiers for long prompts. Model names change often, so read the live page for the model you choose.
Rules that change the bill
Free tier (Gemini). Google marks most Gemini models “Free of charge” in its standard table. Gemini 3.1 Pro Preview is marked not available on the free tier. Free-tier limits and terms are on Google’s page, so read them before relying on it for customer data.
Batch discount. Google, OpenAI and Anthropic each list Batch pricing at half the standard rate for work that does not need an immediate answer, such as overnight document processing.
Caching. OpenAI and Anthropic list cheaper rates for repeated input, and Anthropic lists separate cache-write rates. A workflow that sends the same long instructions on every call can use this. The mechanics differ by vendor.
Price changes already posted. Google’s page lists Gemini 3.8 Flash and Gemini 3.6 Flash at US$0.75 input and US$3.75 output until 31 December 2026, rising to US$1.50 and US$7.50 from 1 January 2027. Read the dates on the page when budgeting for a year.
Long prompts. Google lists a higher rate for Gemini 3.1 Pro Preview prompts above 200k tokens (US$4.00 input, US$18.00 output).
Availability in the Philippines
- Google. The Gemini API and Google AI Studio page lists the Philippines among available countries and territories.
- OpenAI. The Philippines is on OpenAI’s supported-countries list for the API.
- Anthropic. The Philippines is on Anthropic’s supported countries and regions list, which covers Claude.ai, the API and developer platform.
Which payment methods work from the Philippines, and whether a local card is charged in dollars or converted by the bank, was not confirmed from any vendor page. The card issuer sets the foreign-currency fee. Check at checkout.
Getting and keeping an API key
Google’s documentation says to create a Gemini key at aistudio.google.com/apikey, and to “treat your Gemini API key like a password.” It tells developers never to check keys into source control, not to hardcode them in web or mobile apps, to store them in a secret manager in production, and to set billing alerts. The same habits apply to OpenAI and Anthropic keys. A key shared in a chat, a spreadsheet or a public repository can run up charges.
Working out a monthly cost
Estimate the tokens a call uses, multiply by calls per month, and apply the rates above. As arithmetic from the table, a call that sends 1,000 input tokens and returns 500 output tokens on Gemini 2.5 Flash costs US$0.0003 for input and US$0.00125 for output, about US$0.00155 in total. Real calls vary with prompt length, so run a sample of your own documents and read the usage figures the vendor returns. Add the cost of the workflow tool on top. n8n pricing in PHP shows one tool’s side of that sum.
Data handling to check
OpenAI’s developer documentation states that data sent to the API is not used to train its models unless you opt in, and that abuse monitoring logs are kept for up to 30 days by default. Anthropic states that by default it does not use inputs or outputs from commercial products such as the API for training. Google’s terms for the paid and free Gemini API tiers were not read for this page, so check them before sending customer data. Under the Data Privacy Act, a business stays accountable for personal data it passes to a service (Republic Act No. 10173, Section 21). The pillar guide covers the wider rules.
Frequently asked questions
How much does the Gemini API cost?
Google lists Gemini 2.5 Flash at US$0.30 per 1 million input tokens and US$2.50 per 1 million output tokens, about PHP 19 and PHP 157. Most Gemini models also have a free tier. Check the live page.
How do I get a Gemini API key?
Sign in at Google AI Studio, open the API Keys page and choose Create API key. Store the key as a secret and never put it in code that others can read.
Is the OpenAI API available in the Philippines?
Yes. The Philippines is on OpenAI’s supported-countries list for the API.
Is the Claude API available in the Philippines?
Yes. The Philippines is on Anthropic’s supported countries and regions list.
Is the ChatGPT API different from a ChatGPT subscription?
OpenAI’s pricing page lists the API per token on its developer platform. Check the billing terms before assuming a chat subscription covers API calls.
Sources
All sources were read on 12 Oct 2026.
- Google, Gemini API pricing: standard and Batch rates, free tier, dated price changes.
- Google, Gemini API available regions: the Philippines is listed.
- Google, Using Gemini API keys: key creation and handling.
- OpenAI, API pricing: standard and Batch rates.
- OpenAI, Supported countries: the Philippines is listed.
- OpenAI, Your data (API): training default.
- Anthropic, Pricing: per-model rates, Batch discount.
- Anthropic, Supported countries and regions: the Philippines is listed.
- Anthropic Privacy Center, Is my data used for model training?: commercial default.
- Republic Act No. 10173, Data Privacy Act of 2012, via Lawphil.
- European Central Bank euro reference rates, retrieved through the Frankfurter API (api.frankfurter.dev): EUR to PHP and USD, 9 Oct 2026.