API Pricing
Official API rates, context windows, reasoning features, and prompt cache discounts across mainstream LLMs with verified timestamps
Showing 8 API price quotes. Current sort: by provider, release date newest first.
| Model & API ID | Official Source | ||||
|---|---|---|---|---|---|
OpenAI | |||||
Global· gpt-5.6-solView notesEnglish translation pending. Check the official source for pricing conditions. Read original note (Chinese)Standard 标准文本 API；输入不超过 272,000 tokens；更长输入按长上下文档计费；缓存列为读取费用，缓存写入另计。当前官网所列价格；官网注明优惠至少持续至 2026-11-21。 | — | US$ 4.00 Cached Input: US$ 0.40 | US$ 20.00 | ||
Global· gpt-5.6-lunaView notesStandard text API; input up to 272,000 tokens. Longer inputs use long-context pricing. The cached input rate covers cache reads; cache writes are charged separately. | — | US$ 0.20 Cached Input: US$ 0.02 | US$ 1.20 | ||
Global· gpt-6.1-terraView notesStandard text API; input up to 272,000 tokens. Longer inputs use long-context pricing. The cached input rate covers cache reads; cache writes are charged separately. | — | US$ 1.00 Cached Input: US$ 0.05 | US$ 5.00 | ||
Global· gpt-5.6-terraView notesStandard text API; input up to 272,000 tokens. Longer inputs use long-context pricing. The cached input rate covers cache reads; cache writes are charged separately. | — | US$ 2.00 Cached Input: US$ 0.20 | US$ 12.00 | ||
Global· gpt-6-astraView notesStandard text API; input up to 272,000 tokens. Longer inputs use long-context pricing. The cached input rate covers cache reads; cache writes are charged separately. | — | US$ 10.00 Cached Input: US$ 1.00 | US$ 50.00 | ||
Global· gpt-6-lunaView notesStandard text API; input up to 272,000 tokens. Longer inputs use long-context pricing. The cached input rate covers cache reads; cache writes are charged separately. | — | US$ 0.10 Cached Input: US$ 0.01 | US$ 0.50 | ||
Global· gpt-6-solView notesStandard text API; input up to 272,000 tokens. Longer inputs use long-context pricing. The cached input rate covers cache reads; cache writes are charged separately. | — | US$ 2.00 Cached Input: US$ 0.20 | US$ 10.00 | ||
Global· gpt-6.1-solView notesStandard text API; input up to 272,000 tokens. Longer inputs use long-context pricing. The cached input rate covers cache reads; cache writes are charged separately. | — | US$ 2.00 Cached Input: US$ 0.10 | US$ 10.00 | ||
How to read pricing and model specs?
Grouped by provider by default, sorted newest to oldest by public release date. Release date is not price verification date; uncataloged dates appear at the end. Click column headers to sort by price, context window, or verification date across providers. All token prices are standardized per 1M tokens without currency conversion. Context windows not independently audited are marked as unverified. Actual billing is determined by vendor documentation and timestamps.
Price Changes →