Mindket

Prices

AI API prices, kept day by day

What each API charges per million tokens read, written and read from the cache, for the chat models of 11 sellers. Read once a day; every change is kept below with its old price.

Models
241
Sellers
11
Price changes, 30 days
0
Retiring within 90 days
19
Last read

Recent price changes

Each price that moved in the last 90 days, the day Mindket saw it, the old price and the new one (US$ per million tokens).

No price on this page has changed since Mindket began keeping them on 2026-11-13.

Prices that end

Introductory prices with a last day, from the makers’ own price pages (the token calculator uses the same). After that day the price on the right applies.

US$ per million tokens · Input / Output
ModelMakerNowAfterLeft
Gemini 4 ArgonGoogle$2.00 / $10.00$4.00 / $20.00From 2026-12-0117 days
Gemini 3.8 FlashGoogle$0.75 / $3.75$1.50 / $7.50From 2027-01-0148 days

Retiring within 90 days

Models on this page whose seller has named the day it stops serving them.

ModelSellerRetires
gemini-2.5-pro-preview-ttsGoogle Gemini
claude-sonnet-4-5Anthropic
gemini-2.5-flashGoogle Gemini
gemini-2.5-proGoogle Gemini
gemini-omni-1.1-flashGoogle Gemini
gpt-5OpenAI
gpt-5-miniOpenAI
gpt-5-nanoOpenAI
o3OpenAI
gpt-4.1OpenAI
gpt-4.1-miniOpenAI
gpt-4oOpenAI
gpt-4o-miniOpenAI
gpt-4o-audio-preview-2024-12-17OpenAI
gpt-4o-audio-preview-2025-06-03OpenAI
gpt-4o-mini-audio-preview-2024-12-17OpenAI
gpt-audioOpenAI
gpt-audio-miniOpenAI
gpt-5.5-cyber-previewOpenAI

Every AI deadline, as a calendar

OpenAI 43 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
gpt-6-cyber$20.00$100.00$2.00922K——
daybreak-red-latest$12.50$75.00$1.25272K——
gpt-5.5-cyber$12.50$75.00$1.25———
gpt-5.5-cyber-preview$12.50$75.00$1.25——
gpt-5.6-cyber$12.50$75.00$1.25272K——
gpt-6-rosalind$15.00$75.00$1.50922K——
gpt-6-astra$10.00$50.00$1.00922K——
gpt-6-astra-law$10.00$50.00$1.00922K——
chat-latest$5.00$30.00$0.50400K——
gpt-5.5$5.00$30.00$0.501.05M——
gpt-rosalind-discovery$5.00$25.00$0.50———
gpt-rosalind-research$5.00$25.00$0.50———
31 more models
US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
daybreak-blue-latest$4.00$20.00$0.40922K——
gpt-5.6$4.00$20.00$0.40922K——
gpt-5.6-sol$4.00$20.00$0.40922K——
gpt-5.4$2.50$15.00$0.251.05M——
gpt-5.2$1.75$14.00$0.175272K——
gpt-5.6-terra$2.00$12.00$0.20922K——
gpt-4o$2.50$10.00$1.25128K—
gpt-4o-audio-preview-2024-12-17$2.50$10.00—128K—
gpt-4o-audio-preview-2025-06-03$2.50$10.00—128K—
gpt-4o-search-preview$2.50$10.00$1.25128K——
gpt-5$1.25$10.00$0.125272K—
gpt-5-search-api$1.25$10.00$0.125272K——
gpt-5.1$1.25$10.00$0.125272K—
gpt-6-sol$2.00$10.00$0.20922K——
gpt-6.1-sol$2.00$10.00$0.10922K——
gpt-audio$2.50$10.00—128K—
gpt-audio-1.5$2.50$10.00—128K——
gpt-4.1$2.00$8.00$0.501.05M—
o3$2.00$8.00$0.50200K—
gpt-6.1-terra$1.00$5.00$0.05922K——
gpt-5.4-mini$0.75$4.50$0.075272K——
gpt-audio-mini$0.60$2.40—128K—
gpt-5-mini$0.25$2.00$0.025272K—
gpt-4.1-mini$0.40$1.60$0.101.05M—
gpt-5.4-nano$0.20$1.25$0.02272K—
gpt-5.6-luna$0.20$1.20$0.02922K——
gpt-4o-mini$0.15$0.60$0.075128K—
gpt-4o-mini-audio-preview-2024-12-17$0.15$0.60—128K—
gpt-4o-mini-search-preview$0.15$0.60$0.075128K——
gpt-6-luna$0.10$0.50$0.01922K——
gpt-5-nano$0.05$0.40$0.005272K—

Anthropic 18 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
claude-fable-5$10.00$50.00$1.001M——
claude-fable-5-1$10.00$50.00$0.251M——
claude-mythos-5$10.00$50.00$1.001M——
claude-mythos-5-1$10.00$50.00$0.251M——
claude-fable-5-2$8.00$40.00$0.201M——
claude-mythos-5-2$8.00$40.00$0.201M——
claude-opus-4-5$5.00$25.00$0.50200K——
claude-opus-4-6$5.00$25.00$0.501M——
claude-opus-4-7$5.00$25.00$0.501M——
claude-opus-4-8$5.00$25.00$0.501M——
claude-opus-5$5.00$25.00$0.501M——
claude-opus-5-5$4.00$20.00$0.201M——
6 more models
US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
claude-sonnet-4-5$3.00$15.00$0.30200K—
claude-sonnet-4-6$3.00$15.00$0.301M——
claude-sonnet-5$2.00$10.00$0.201M——
claude-sonnet-5-5$2.00$10.00$0.101M——
claude-haiku-4-5$1.00$5.00$0.10200K——
claude-haiku-5-5$0.10$0.50$0.011M——

Google Gemini 39 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
gemini-2.5-pro-preview-tts$1.00$20.00$0.1258K—
deep-research-max-preview-04-2026$2.00$12.00—1.05M——
deep-research-preview-04-2026$2.00$12.00—1.05M——
gemini-3.1-pro-preview$2.00$12.00$0.201.05M——
gemini-3.1-pro-preview-customtools$2.00$12.00$0.201.05M——
gemini-pro-latest$2.00$12.00$0.201.05M——
gemini-2.5-pro$1.25$10.00$0.1251.05M—
gemini-4-argon$2.00$10.00$0.101.05M——
gemini-4-carbon$2.00$10.00$0.101.05M——
gemini-3.5-flash$1.50$9.00$0.151.05M——
gemini-omni-1.1-flash$1.50$9.00—131K—
gemini-omni-1.2-flash$1.50$9.00—131K——
27 more models
US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
gemini-4-hydrogen$1.50$7.50$0.0751.05M——
gemini-robotics-er-2-preview$1.00$5.00$0.10131K——
gemini-robotics-er-2-streaming-preview$1.00$5.00—131K——
gemini-3.6-flash$0.75$3.75$0.0751.05M——
gemini-3.7-flash$0.75$3.75$0.0751.05M——
gemini-3.8-flash$0.75$3.75$0.0751.05M——
gemini-flash-latest$0.75$3.75$0.0751.05M——
gemini-3-flash-preview$0.50$3.00$0.051.05M——
gemini-2.5-flash$0.30$2.50$0.031.05M—
gemini-3.5-flash-lite$0.30$2.50$0.031.05M——
gemini-flash-lite-latest$0.30$2.50$0.031.05M——
gemini-3.1-flash-lite$0.25$1.50$0.0251.05M—
gemini-gemma-2-27b-it$0.35$1.05—8K——
gemini-gemma-2-9b-it$0.35$1.05—8K——
gemini-2.5-flash-lite$0.10$0.40$0.011.05M——
gemini-exp-1114$0.00$0.00—1.05M——
gemini-exp-1206$0.00$0.00—2.1M——
gemma-3-27b-it$0.00$0.00—131K——
gemma-4-26b-a4b-it$0.00$0.00—262K——
gemma-4-31b-it$0.00$0.00—262K——
learnlm-1.5-pro-experimental$0.00$0.00—33K——
lyria-3-clip-preview$0.00$0.00—131K——
lyria-3-pro-preview$0.00$0.00—131K——
lyria-3.5$0.00$0.00—131K——
lyria-3.5-clip-preview$0.00$0.00—131K——
lyria-3.5-pro-preview$0.00$0.00—131K——
lyria-realtime-exp$0.00$0.00—1.05M——

xAI 29 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
grok-4.5$2.00$6.00$0.30500K——
grok-4.5-latest$2.00$6.00$0.30500K——
grok-4.6$2.00$6.00$0.50500K——
grok-4.7$2.00$6.00$0.50500K——
grok-build-latest$2.00$6.00$0.30500K——
grok-4.20$1.25$2.50$0.201M——
grok-4.20-beta$1.25$2.50$0.201M——
grok-4.20-beta-latest$1.25$2.50$0.201M——
grok-4.20-beta-latest-non-reasoning$1.25$2.50$0.201M——
grok-4.20-beta-latest-reasoning$1.25$2.50$0.201M——
grok-4.20-beta-non-reasoning$1.25$2.50$0.201M——
grok-4.20-beta-reasoning$1.25$2.50$0.201M——
17 more models
US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
grok-4.20-experimental-beta-0304$1.25$2.50$0.201M——
grok-4.20-experimental-beta-0304-non-reasoning$1.25$2.50$0.201M——
grok-4.20-experimental-beta-0304-reasoning$1.25$2.50$0.201M——
grok-4.20-experimental-beta-latest$1.25$2.50$0.201M——
grok-4.20-experimental-beta-non-reasoning-latest$1.25$2.50$0.201M——
grok-4.20-experimental-beta-reasoning-latest$1.25$2.50$0.201M——
grok-4.20-non-reasoning$1.25$2.50$0.201M——
grok-4.20-non-reasoning-gv2$1.25$2.50$0.201M——
grok-4.20-non-reasoning-latest$1.25$2.50$0.201M——
grok-4.20-reasoning$1.25$2.50$0.201M——
grok-4.20-reasoning-gv2$1.25$2.50$0.201M——
grok-4.20-reasoning-latest$1.25$2.50$0.201M——
grok-4.3$1.25$2.50$0.201M——
grok-4.3-latest$1.25$2.50$0.201M——
grok-build-0.1$1.00$2.00$0.20256K——
grok-code-fast$1.00$2.00$0.20256K——
grok-code-fast-1-0825$1.00$2.00$0.20256K——

DeepSeek 8 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
deepseek-v4-pro$1.32$3.96$0.0441M——
deepseek-r1$0.55$2.19$0.1466K——
deepseek-flash$0.30$1.20$0.0061M——
deepseek-v4-flash$0.30$1.20$0.0061M——
deepseek-v4-flash-vision-exp$0.30$1.20$0.0061M——
deepseek-v3$0.27$1.10$0.0766K——
deepseek-v3.2$0.28$0.40$0.028164K——
deepseek-coder$0.14$0.28$0.014128K——

Mistral AI 47 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
magistral-medium-latest$1.50$7.50$0.15262K——
mistral-medium$1.50$7.50$0.15262K——
mistral-medium-3$1.50$7.50$0.15262K——
mistral-medium-3-5$1.50$7.50$0.15262K—
mistral-medium-3.5$1.50$7.50$0.15262K——
mistral-medium-latest$1.50$7.50$0.15262K——
mistral-vibe-cli-latest$1.50$7.50$0.15262K——
mistral-vibe-cli-with-tools$1.50$7.50$0.15262K——
pixtral-large-latest$2.00$6.00$0.20128K——
glm-5-2$1.40$4.40$0.141.05M——
zai-glm-5$1.40$4.40$0.141.05M——
zai-glm-5-3$1.40$4.40$0.141.05M——
35 more models
US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
zai-glm-latest$1.40$4.40$0.141.05M——
mistral-large-4$0.70$2.10$0.071.05M——
mistral-large-4-0$0.70$2.10$0.071.05M——
devstral-latest$0.40$2.00$0.04256K——
devstral-medium-latest$0.40$2.00$0.04256K——
mistral-code-agent-latest$0.40$2.00$0.04256K——
mistral-large-2512$0.50$1.50$0.05262K—
mistral-large-3$0.50$1.50$0.05262K——
mistral-large-latest$0.50$1.50$0.05262K——
deepseek-v4-1-flash-1017$0.30$1.20$0.0071.05M——
codestral-2508$0.30$0.90$0.03128K——
codestral-latest$0.30$0.90$0.03128K——
mistral-code-fim-latest$0.30$0.90$0.03128K——
mistral-code-latest$0.30$0.90$0.03128K——
magistral-small-latest$0.15$0.60$0.015262K——
mistral-small-latest$0.15$0.60$0.015262K——
mistral-vibe-cli-fast$0.15$0.60$0.015262K——
zai-glm-5-3-flash$0.15$0.50$0.031.05M——
voxtral-small-2507$0.10$0.40$0.0133K——
voxtral-small-latest$0.10$0.40$0.0133K——
devstral-small-latest$0.10$0.30$0.01256K——
mistral-small$0.10$0.30$0.0132K——
open-mistral-nemo$0.30$0.30$0.03128K——
codestral-mamba-latest$0.25$0.25$0.025256K——
mistral-tiny$0.25$0.25$0.02532K——
ministral-14b-2512$0.20$0.20$0.02262K——
ministral-14b-latest$0.20$0.20$0.02262K——
ministral-3-14b-2512$0.20$0.20$0.02262K——
ministral-3-8b-2512$0.15$0.15$0.015262K——
ministral-8b-2512$0.15$0.15$0.015262K——
ministral-8b-latest$0.15$0.15$0.015262K——
ministral-3-3b-2512$0.10$0.10$0.01131K——
ministral-3b-2512$0.10$0.10$0.01131K——
ministral-3b-latest$0.10$0.10$0.01131K——
labs-leanstral-1-5-1$0.00$0.00—262K——

Moonshot AI (Kimi) 5 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
kimi-k3.1-pro-highspeed$4.00$20.00$0.201.05M——
kimi-k3$3.00$15.00$0.301.05M——
kimi-k3.1-pro$2.00$10.00$0.201.05M——
kimi-k2.6$0.95$4.00$0.16262K——
kimi-k2.7-code$0.95$4.00$0.19262K——

Alibaba Cloud (Qwen) 21 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
qwen3.7-max$2.50$7.50$0.50992K——
qwen-max$1.60$6.40—31K——
qwen3.8-max$2.00$6.00$0.25992K——
deepseek-v4-pro$2.40$4.80$0.201M——
glm-5.1$1.40$4.40$0.26203K——
glm-5.2$1.40$4.40$0.281.05M——
kimi-k2.7-code$0.95$4.00$0.19229K——
qwen3-vl-235b-a22b-thinking$0.40$4.00—131K——
qwen3-vl-32b-thinking$0.16$2.87—131K——
qwq-plus$0.80$2.40—98K——
qwen3-vl-235b-a22b-instruct$0.40$1.60—131K——
qwen-coder$0.30$1.50—1M——
9 more models
US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
qwen-plus$0.40$1.20—129K——
qwen3-next-80b-a3b-instruct$0.15$1.20—262K——
qwen3-next-80b-a3b-thinking$0.15$1.20—262K——
qwen3-vl-32b-instruct$0.16$0.64—131K——
qwen3.8-flash$0.15$0.47$0.016992K——
qwen3.8-omni-flash$0.15$0.47$0.016992K——
deepseek-v4-flash$0.20$0.40$0.041M——
qwen-turbo$0.05$0.20—129K——
qwen-turbo-latest$0.05$0.20—1M——

Z.ai (GLM) 17 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
glm-4.5-x$2.20$8.90—128K——
glm-5-code$1.20$5.00$0.30200K——
glm-4.5-airx$1.10$4.50—128K——
glm-5.1$1.40$4.40$0.26200K——
glm-5.2$1.40$4.40$0.261M——
glm-5.3$1.40$4.40$0.261M——
glm-5v-turbo$1.20$4.00$0.24200K——
glm-5$1.00$3.20$0.20200K——
glm-4.5$0.60$2.20—128K——
glm-4.6$0.60$2.20$0.11200K——
glm-4.7$0.60$2.20$0.11200K——
glm-4.5v$0.60$1.80—128K——
5 more models
US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
glm-4.5-air$0.20$1.10—128K——
glm-5.3-flash$0.15$0.50$0.031.05M——
glm-4-32b-0414-128k$0.10$0.10—128K——
glm-4.5-flash$0.00$0.00—128K——
glm-4.7-flash$0.00$0.00$0.00200K——

Volcano Engine (Doubao) 7 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
doubao-seed-2-1-pro-260628$0.8625$4.31$0.1725256K——
doubao-seed-2-1-pro-260915$0.8625$4.31$0.17251.02M——
doubao-seed-2-1-turbo-260628$0.43125$2.16$0.08625256K——
doubao-seed-2-1-lite-260915$0.115$0.388126$0.0231.02M——
deepseek-v3-2-251201$0.00$0.00—98K——
glm-4-7-251222$0.00$0.00—205K——
kimi-k2-thinking-251104$0.00$0.00—229K——

MiniMax 7 models

US$ per million tokens
ModelInputOutputCached inputContextLast changeRetires
MiniMax-M2.1-lightning$0.30$2.40$0.031M——
MiniMax-M2.5-lightning$0.30$2.40$0.031M——
MiniMax-M2$0.30$1.20$0.03200K——
MiniMax-M2.1$0.30$1.20$0.031M——
MiniMax-M2.5$0.30$1.20$0.031M——
MiniMax-M3$0.30$1.20$0.061M——
MiniMax-M3.1-Flash$0.15$0.60$0.0151M——

Claude’s own plans

A flat price a month instead of a price per token, with usage limits that reset every five hours and every week:

  • Pro$20 a month · $17 a month billed yearly
  • Max$100 a month · 5× Pro’s usage · $100 of API credits a month
  • Ultra$200 a month · 10× Pro’s usage · no five-hour limit · $200 of API credits a month

Read on 2026-11-09; prices before tax. claude.com/pricing

Spend fewer tokens for the same work

Claude Token Freedom Handbook: A step-by-step member area for Claude Code in VS Code on a Claude Pro or Max plan, plus the playbook's ten files.

See the playbook

How this is made

The prices come from LiteLLM’s model price file, an open list of what each API charges per token. Mindket reads it once a day and shows the chat models of the 11 sellers above, each price per million tokens.

A model with a dated name is left out when the same model is listed without the date. A model past its retirement day is left out.

Each day’s read is compared with the last price kept for each model: a change is kept with its old price, so the history starts on the day Mindket first read the file.

Read on 2026-11-13.

Price data: LiteLLM, MIT License, Copyright (c) 2023 Berri AI (the price file, the license)

Download every price as CSV

Questions

Where do these prices come from?

From LiteLLM’s model price file, an open list (MIT license) of what each AI API charges per token, read by Mindket once a day. This page shows the chat models of 11 sellers, per million tokens, and keeps every price that changes.

Which model here costs least per million output tokens?

ministral-3-3b-2512 from Mistral AI: $0.10 per million tokens written and $0.10 per million read.

Which prices are about to change?

Gemini 4 Argon’s introductory price ends on 2026-11-30; then $4.00 read and $20.00 written per million tokens. Gemini 3.8 Flash’s introductory price ends on 2026-12-31; then $1.50 read and $7.50 written per million tokens.

How do I see when a price changed?

Recent price changes lists each change with its day, the old price and the new one, and each model’s table shows its last change. The CSV holds the prices as of the last read.