GPT-3.5 Turbo API Pricing
GPT-3.5 Turbo has 7 API providers. The lowest input price is $0.5000/1M tokens from OpenRouter. 7 providers auto-refresh every 6 h (Updated 1h ago).
No first-party price is tracked for GPT-3.5 Turbo on this page yet — below is a comparison of 7 hosting providers: prices range from $0.5000 to $12.0000, a 2300% spread. Beyond price, hosts differ meaningfully in throughput, rate limits, and reliability, so a small trial run is worth it before committing.
GPT-3.5 Turbo supports a 16K context window. At today's lowest rate, processing 1M input + 1M output tokens costs about $2.0000. The table below lists current per-provider quotes.
Compare GPT-3.5 Turbo API Prices Across Providers
Frequently Asked Questions
How is GPT-3.5 Turbo API pricing calculated?
LLM APIs are billed per 1M input and output tokens separately. Official providers set the base price; relay providers typically offer 20–80% discounts.
What is the cheapest way to access GPT-3.5 Turbo API?
The lowest current input price is $0.5000 per 1M tokens from OpenRouter. Prices update in real time — bookmark this page for the latest rates.
Can I use GPT-3.5 Turbo API from China?
Providers marked 🇨🇳 in the table support China-mainland access. Check each provider's documentation for details.
How often is GPT-3.5 Turbo API pricing updated?
7 sources on this page are live-scraped every ~6 hours (Updated 1h ago).
How reliable is the GPT-3.5 Turbo API pricing data?
Live-scraped prices come directly from provider APIs and are generally accurate. Manually maintained prices are sourced from official pricing pages and periodically verified. For critical decisions, confirm the latest price via the provider's official link.
How can I tell if a price is live-scraped or manually maintained?
Each provider row shows a small label next to the name: a green "Live" badge means prices are automatically fetched on a schedule; a gray "Manual" badge means human-curated. Live data not refreshed in 24+ hours turns amber.
Related