1M-token context API pricing: the cheapest long-context models at every provider

Updated Oct 31, 2026

A long context lets a model read a whole codebase, a contract bundle or a book in one request. A million tokens is roughly 750,000 words. Dozens of models now accept that much, and the price of reading it differs more than tenfold. Below: the models with a context of 1M tokens or more that at least five providers sell, with the typical price (the median across the providers that offer the full context). All of them in one list: models with 1M+ context, cheapest first.

The cheapest 1M-token models

Price per 1M tokens, input / output, in the 30 October 2026 collection, median across providers that sell the full context:

• Qwen3.7 Flash (Alibaba) - $0.03 / $0.13, 11 providers. The cheapest long context in the catalog.
• Qwen3.5 Flash - $0.065 / $0.28, 9 providers.
• Gemini 2.5 Flash-Lite (Google) - $0.10 / $0.40, input from $0.05; 16 providers.
• GPT-4.1 nano (OpenAI) - $0.10 / $0.40, 16 providers.
• Claude Haiku 5.5 (Anthropic) - $0.10 / $0.50, 20 providers.
• GPT-6 Luna (OpenAI) - $0.10 / $0.50, 19 providers.
• DeepSeek V4 Flash - $0.14 / $0.30, 24 providers.
• GLM-5.3 Flash (Zhipu) - $0.15 / $0.50, 1.3M context, 43 providers - the most widely sold.
• Grok 4.1 Fast (xAI) - $0.20 / $0.50, 2M context, 6 providers.

A step up, $0.25-0.30 input: Gemini 3.1 Flash-Lite ($0.25 / $1.50), Gemini 2.5 Flash ($0.30 / $2.50), MiniMax M3 ($0.30 / $1.20), DeepSeek V4.1 Flash ($0.30 / $1.20).

What one full prompt costs

Reading 1M tokens once costs the input price: 3 cents with Qwen3.7 Flash, 10 cents with Gemini 2.5 Flash-Lite or Claude Haiku 5.5, 20 cents with Grok 4.1 Fast, 30 cents with Gemini 2.5 Flash. A hundred such requests a day for a month (3,000M tokens) is $90 against $900 - the choice of model decides the bill.

Grok 4.1 Fast: every provider

Live data: this table is built from today's prices.

Gemini 2.5 Flash-Lite: every provider

Live data: this table is built from today's prices.

DeepSeek V4.1 Flash: every provider

Live data: this table is built from today's prices.

Before you choose

• Not every provider sells the full context. The same model can be capped at 128K or 256K by a reseller - check the context column on the card.
• Long prompts can cost more per token. Some providers raise the price above 200K tokens; the card shows such tiers as a condition.
• Caching helps. If you send the same long document again and again, a cached-input price cuts the cost several times - many cards list it.
• Quality over length: a model that reads 1M tokens does not use all of it equally well - test on your own documents.

All prices are in USD per 1 million tokens, without taxes, as the providers publish them; each row links to its card with the source and the date it was checked.