AI API price changes
Newest effective date first. Each row is a change to a stored list price, a model added to the ledger, or a scheduled rate an official page has already published. Rates are USD per 1M tokens, shown as input / cached input / output.
· Scheduled change · Google
Gemini 3.8 Flash
$0.75 / $0.075 / $3.75 → $1.50 / $0.15 / $7.50
The Gemini API pricing page schedules Gemini 3.8 Flash's paid standard rate to move from $0.75 input, $0.075 cached input, and $3.75 output per 1M tokens to $1.50, $0.15, and $7.50 on 1 January 2027.
· Scheduled change · Google
Gemini 4 Argon
$2.00 / $0.10 / $10.00 → $4.00 / $0.20 / $20.00
The Gemini API pricing page schedules Gemini 4 Argon's paid standard rate to move from $2 input, $0.10 cached input, and $10 output per 1M tokens to $4, $0.20, and $20 on 1 December 2026.
· Price cut · Mistral
Mistral Large 4
$1.36 / $0.14 / $4.18 → $0.70 / $0.07 / $2.10
Mistral's 29 October 2026 announcement set Mistral Large 4 at $0.70 input, $0.07 cached input, and $2.10 output per 1M tokens, down from the previous ledger rate of $1.36, $0.14, and $4.18, and the API pricing table lists the same rates; the ledger recorded this change on 30 October 2026.
· New model · Google
Gemini 4 Carbon
n/a → $2.00 / $0.10 / $10.00
Google announced Gemini 4 Carbon on 28 October 2026 with access limited to Fairwind Program partners and an introductory rate of $2 input, $0.10 cached input, and $10 output per 1M tokens.