Skip to content
Promtime

Guide

AI prices

What subscriptions and APIs cost across companies

Updated 21 October 2026139 stories

In short

On the API side, OpenAI’s GPT-6.1 Sol lists at $2/$10 per million input/output tokens, with cached input at $0.10 per million. Anthropic’s Claude Sonnet 5.5 is $2/$10 per Mtok with 1M context, and its cache reads were halved from $0.20 to $0.10 per million tokens. Google’s Gemini 4 Argon costs an introductory $2 input and $10 output per million tokens through November 30, then $4 and $20. At the low end, Claude Haiku 5.5 charges $0.10/$0.50 for prompts up to 100K tokens and MiniMax M3.1 Flash charges $0.15 input and $0.60 output. Speed costs extra: Ultrafast on GPT-6.1 Sol is $12/$60.

On subscriptions, ChatGPT Work and Codex usage on OpenAI’s $200 Pro plan falls from 20x to 10x the Plus allowance on October 30, and a new $500 Pro plan gives 25x. xAI’s The X Subscription runs from Plus at $8/month to Ultra at $200, and its $30 Premium tier bundles SuperGrok, Cursor Pro and X Premium+.

Latest change: OpenAI now keeps the cache forever for logged-in Codex and Work users and gives API developers a 12-hour cache, so resuming an old conversation carries no penalty for those users. On the Anthropic side, Claude Code 2.1.304 on Pro and Max asks first before you resume an inactive conversation, because that can use a noticeable share of the 5-hour limit. DeepSeek web search costs $0.75 per 1,000 successful requests, and server-side calls through DeepSeek Harness, Claude Code and Codex are free until October 31.

At a glance

CompanyPlan or APIPriceAs ofSource
OpenAIGPT-6.1 Sol (API)$2/$10 per million input/output tokens; cached input $0.10 per million29 September 2026Read
OpenAIGPT-6.1 Sol Ultrafast mode (ChatGPT Work, Codex, API)$12 per million input tokens, $60 per million output tokens8 October 2026Read
OpenAIGPT-6 Astra (API)$10 in, $50 out4 September 2026Read
OpenAIPro $200 plan$200; from October 30, ChatGPT Work and Codex usage falls from 20x to 10x the Plus allowance, and GPT-6 Pro in Chat drops from 200 to 100 messages per week30 September 2026Read
OpenAIPro $500 plan$500; 25x Plus and Ultrafast for GPT-6 Astra30 September 2026Read
OpenAIChatGPT Business Premium seats$100 a month14 August 2026Read
AnthropicClaude Sonnet 5.5 (Anthropic API)$2/$10 per Mtok, 1M context28 September 2026Read
AnthropicClaude Sonnet 5.5 cache reads$0.10 per million tokens (was $0.20)7 October 2026Read
AnthropicClaude Haiku 5.5 (claude-haiku-5-5)$0.10/$0.50 per million input/output tokens for prompts up to 100K tokens; $0.50/$2.50 above 100K7 October 2026Read
AnthropicMonthly API credits on Claude Max and TeamMax 5x $100 a month, Max 20x $200; Team $20 per Standard seat and $100 per Premium seat, one shared balance capped at $500; Pro gets nothing7 October 2026Read
GoogleGemini 4 Argon (paid tier of the Gemini API)$2 per million input tokens and $10 per million output through November 30, cached input 95% off; then $4 and $2013 October 2026Read
GoogleNano Banana 2.1$30 per 1,000,000 output tokens; $0.076 per 4096x4096px image6 October 2026Read
Show all (21)
CompanyPlan or APIPriceAs ofSource
xAIThe X Subscription Plus$8/month13 October 2026Read
xAIThe X Subscription Premium (SuperGrok, Cursor Pro, X Premium+)$30/month13 October 2026Read
xAIThe X Subscription Ultra$20013 October 2026Read
CursorPremium plan$30; Grok from an included weekly pool, every Claude request billed on top at API list prices13 October 2026Read
DeepSeekSearch and Fetch APIs$0.75 per 1,000 successful requests; server-side calls through DeepSeek Harness, Claude Code and Codex free until October 3119 October 2026Read
MiniMaxM3.1 Flash (MiniMax API)$0.15 input, $0.60 output, $0.015 cached input per 1M tokens19 October 2026Read
MicrosoftMicrosoft-Decision-1 (Microsoft Foundry)$0.042 per million input tokens; output tokens cost nothing9 October 2026Read
falFLUX 3 Max$0.0425 per second at 720p and $0.0725 at 1080p through November 7, then $0.17 and $0.2919 October 2026Read
KreaKrea 3about $0.06 per 2K edit (Krea's side-by-side)15 October 2026Read

Milestones

  1. Claude Fable 5 becomes a permanent part of Max and Team Premium.
  2. Claude Opus 5 launches at Opus 4.8 pricing and becomes the Max default.
  3. Sonnet 5’s introductory pricing becomes permanent.
  4. GPT-5.6 Luna output drops from $6 to $1.20 per million tokens.
  5. Cache reads get 75% cheaper with Claude Fable 5.1.
  6. GPT-6 Astra lands on the API at $10 in, $50 out.
  7. OpenAI pauses new $200 Pro signups as Astra demand spikes.
  8. Opus 5.5 becomes the default model in Claude Code.
  9. Claude Sonnet 5.5 becomes the default Sonnet on the Anthropic API at $2/$10 per Mtok with 1M context.
  10. GPT-6.1 Sol ships at $2/$10 per million tokens, with cached input cut to $0.10 per million.
  11. OpenAI halves Codex and Work usage on the $200 Pro plan and adds a $500 Pro tier at 25x Plus.
  12. Claude Haiku 5.5 launches at $0.10/$0.50 per million tokens for prompts up to 100K tokens.
  13. Claude Max and Team plans get monthly API credits that don’t cover interactive Claude Code sessions.
  14. Ultrafast mode comes to GPT-6.1 Sol at $12/$60 per million tokens.
  15. xAI launches The X Subscription, which folds SuperGrok, Cursor Pro and X Premium+ into a $30 Premium tier.
  16. Gemini 4 Argon launches on the paid Gemini API at an introductory $2/$10 per million tokens.
  17. DeepSeek opens its web search and fetch APIs at $0.75 per 1,000 successful requests.
  18. OpenAI keeps the cache forever for Codex and Work users and gives API developers a 12-hour cache.

FAQ

How much does GPT-6.1 Sol cost in the API?

It’s $2/$10 per million input/output tokens, with cached input at $0.10 per million, which is 50% less than GPT-6 Sol’s cached rate.

Ultrafast mode, which OpenAI says is eight times faster than Sol Standard, costs $12 per million input tokens and $60 per million output tokens. In the Responses API you turn it on with service_tier: “ultrafast”.

What changes on the ChatGPT Pro $200 plan on October 30?

ChatGPT Work and Codex usage falls from 20x to 10x the Plus allowance, and GPT-6 Pro in Chat drops from 200 to 100 messages per week. Current limits hold through October 29. After that, existing subscribers get 62,500 usage credits, which OpenAI values at $2,500 and which expire December 31, 2026.

A new $500 Pro plan gives 25x Plus, plus Ultrafast for GPT-6 Astra.

How much does Claude Haiku 5.5 cost?

For prompts up to 100K tokens it’s $0.10/$0.50 per million input/output tokens, compared with $1/$5 for Haiku 4.5. Above 100K tokens the rate rises to $0.50/$2.50. Anthropic puts average savings at around 75%, because an updated tokenizer uses slightly more tokens per task.

What does Gemini 4 Argon cost in the API?

Through November 30, Google charges an introductory $2 per million input tokens and $10 per million output tokens, with cached input 95% off. After that the price rises to $4 and $20. The 1 million tokens are a single limit shared by input and output.

What does xAI’s $30 X Subscription Premium include?

Premium bundles SuperGrok, Cursor Pro and X Premium+, which xAI says previously cost $90/month combined. All paid tiers share one pool of Grok model usage across the Grok apps, the Grok tab on X and Cursor.

In Cursor on the $30 Premium plan, Grok usage comes from an included weekly pool, and every Claude request is billed on top at API list prices.

Latest