AI Token Cost Calculator

Selected text-token prices checked 2026-10-22. USD only.

Standard global endpoints; uncached input unless the entered tokens are cache hits. DeepSeek defaults to peak. Excludes cache writes/storage, media, tools, taxes and regional premiums. Enter a per-request scenario; tokenizers differ across providers.

Request scenario

Use observed cache hits. Turning caching on does not guarantee a hit; writes and storage are additional charges.

Models in this pricing snapshot

OpenAI

Anthropic

Google

Mistral

xAI

DeepSeek

Estimated token charges

GPT-5.6 Sol

Unavailable

Checking the price snapshot date...

OpenAI pricing source · checked 2026-10-22

Promotional pricing is stated to last at least through 21 November 2026. Long-context cache-hit pricing is not verified here.

Gemini 3.8 Flash

Unavailable

Checking the price snapshot date...

Google pricing source · checked 2026-10-22

Published prices rise on 1 January 2027; this snapshot stops estimating after 31 December until rechecked. Cache storage is extra.

Grok 4.6

Unavailable

Checking the price snapshot date...

xAI pricing source · checked 2026-10-22

DeepSeek V4 Pro (0813)

Unavailable

Checking the price snapshot date...

DeepSeek pricing source · checked 2026-10-22

Default is peak. Choose the billing window manually: peak is 01:00–04:00 and 06:00–10:00 UTC on weekdays except Chinese public holidays; other times are off-peak.

Recorded base prices

Snapshot checked 2026-10-22, USD per million text tokens. These are standard short-context rates; DeepSeek uses peak prices. Long-context, off-peak and additional service charges can differ. View price data as JSON.

ModelUncached inputCache-hit inputOutputSource
GPT-6 Astra$10$1$50OpenAI
GPT-5.6 Sol$4$0.4$20OpenAI
Claude Fable 5.1$10$0.25$50Anthropic
Claude Fable 5$10$1$50Anthropic
Claude Opus 5$5$0.5$25Anthropic
Claude Sonnet 5$2$0.2$10Anthropic
Claude Opus 4.8$5$0.5$25Anthropic
Claude Opus 4.7$5$0.5$25Anthropic
Claude Sonnet 4.6$3$0.3$15Anthropic
Claude Haiku 4.5$1$0.1$5Anthropic
Gemini 3.8 Flash$0.75$0.075$3.75Google
Gemini 3.1 Pro Preview$2$0.2$12Google
Gemini 3.1 Flash-Lite$0.25$0.025$1.5Google
Mistral Large 3$0.5$0.05$1.5Mistral
Mistral Small 4$0.15$0.015$0.6Mistral
Grok 4.6$2$0.5$6xAI
Grok 4.5$2$0.3$6xAI
DeepSeek V4 Pro (0813)$1.32$0.044$3.96DeepSeek

What this estimate covers

This estimates the entered text-token components at standard global endpoint rates. It does not validate model context/output limits or predict task quality. Use provider usage counters, including billed reasoning tokens. No universal image-token allowance, exchange rate or cache discount is assumed.

Some legacy quotes are withheld pending verification: Unverified legacy OpenAI Terra/Luna/5.5 quotes; Gemini 3.5 Flash and legacy 3 Flash quotes; Retired DeepSeek V4 Flash alias (now routed to a different model); Unverified Kimi and Meta API quotes. This is a selected pricing snapshot, not a complete current-model catalog.