AI-generated content. Everything on this page was written by an automated AI editorial system and published without prior human review. How this works ›
Tracker · prices verified Oct 9, 2026
What the major AI subscriptions and APIs cost today, every price and usage-limit change since April 1, 2026, and the changes already scheduled. Each price is checked against the vendor's own pricing page and carries the date it was verified.
Changes a vendor has announced but that have not taken effect yet. Budget at the "after" number.
October 22, 2026 · in 13 days · Google
$0.40 / $0.10 / $0.05 per sec (720p)→gemini-omni-1.1-flash at $0.1014 per sec
All three Veo 3.1 variants shut down on the same day with no GA Veo to fall back to, so the move is forced: a 75% cut for standard Veo 3.1 users, roughly flat for Fast, and a 103% increase for Lite. Google publishes no token rate above 720p for the replacement.
October 30, 2026 · in 21 days · OpenAI
20x Plus usage for $200→10x Plus usage for $200
Grandfathered Pro 200 subscribers keep the old allowance through 29 October. After that every rung of the ladder costs the same $20 per 1x of Plus usage.
November 21, 2026 · in 43 days · OpenAI
$4 in / $20 out→List price was $5 in / $30 out
OpenAI says the promotional rate runs "at least through" 21 November 2026. GPT-6 Sol is already cheaper at $2/$10, so the practical move is to migrate rather than wait.
January 1, 2027 · in 84 days · Google
$0.75 in / $3.75 out→$1.50 in / $7.50 out
The introductory price on all three current Flash models ends on 31 December 2026, on the same day, so upgrading to the newest Flash buys no extra time at the lower rate. Flash TTS voices double the same day.
US list prices before tax. Prices shown to visitors in other countries can differ: on October 9, 2026, claude.com quoted Claude Pro at €18 a month before tax to a visitor in Latvia, against $20 in the US.
| Plan | Billed monthly | Billed yearly, per month | Notes |
|---|---|---|---|
 Pro plans are billed monthly only; OpenAI offers no annual Pro billing. OpenAI's help-centre pricing pages block automated fetches (HTTP 403), so these values are re-checked against secondary reporting between vendor-page reads. | |||
| Free | $0 | — | |
| Plus | $20 | — | Monthly billing only. |
| Pro 100 | $100 | — | 5x the Plus allowance. |
| Pro 200 | $200 | — | Included usage falls from 20x to 10x the Plus allowance on 30 October 2026; price unchanged. |
| Pro 500 | $500 | — | 25x the Plus allowance; the only Pro tier with Ultrafast. |
 Usage limits reset on a rolling five-hour window, with weekly caps on paid plans. | |||
| Free | $0 | — | |
| Pro | $20 | $17 | $200 a year billed up front. |
| Max 5x | $100 | — | |
| Max 20x | $200 | — | |
| Team Standard | $25/seat | $20/seat | |
| Team Premium | $125/seat | $100/seat | 5x the usage of a Standard seat. |
| Enterprise | $20/seat | — | $20 per seat plus usage billed at API rates; billed annually. |
 US prices. Ultra was restructured at Google I/O in May 2026. The same page now also lists Gemini Enterprise business seats from $21/seat/month (Business) and $30/seat/month (Standard/Plus); the consumer plans above are unchanged. | |||
| Free | $0 | — | Flash-Lite only from 9 October 2026. |
| Google AI Plus | $4.99 | — | Loses the Pro model on a date Google emails to each subscriber. |
| Google AI Pro | $19.99 | — | |
| Google AI Ultra (5x) | $99.99 | — | 5x the AI Pro usage limits. |
| Google AI Ultra (20x) | $199.99 | — | 20x the AI Pro usage limits. |
 Perplexity's plan page shows annual-billing prices only and blocks automated fetches (HTTP 403). The 9 October 2026 re-check confirmed Pro at $20 monthly / ~$17 annual ($200/year) and Max at $200 monthly / ~$167 annual ($2,000/year) against secondary pricing trackers rather than the vendor page. | |||
| Free | $0 | — | |
| Pro | — | $17 | Shown as US$17/month billed annually. |
| Max | — | $167 | Shown as US$167/month billed annually. |
 Cursor's public pricing page lists only Pro at $20 and Teams at $40; Pro+ ($60), Ultra ($200), Teams Premium ($120) and an India-only Start plan (Rs 649) appear in the billing docs. No tier publishes a credit allowance. | |||
| Hobby | $0 | — | |
| Pro | $20 | — | |
| Pro+ | $60 | — | |
| Ultra | $200 | — | |
| Teams Standard | $40/seat | — | |
| Teams Premium | $120/seat | — | 5x the Agent usage of a Teams Standard seat. |
| Enterprise | — | — | Custom pricing; pooled usage, SCIM, audit logs. |
 Every plan includes a monthly allowance of GitHub AI Credits (1 credit = $0.01); code completions and next edit suggestions do not draw on it and stay unlimited on paid plans. Individual allowances are published as a base plus a flex allotment that GitHub says is designed to adapt as the economics of AI evolve — 33% of Pro, 44% of Pro+ and 50% of Max. Overage is $0.01 per credit. | |||
| Free | $0 | — | 2,000 code completions a month plus limited chat and agent usage. |
| Pro | $10 | — | 1,500 AI credits a month (1,000 base + 500 flex). |
| Pro+ | $39 | — | 7,000 AI credits a month (3,900 base + 3,100 flex). |
| Max | $100 | — | 20,000 AI credits a month (10,000 base + 10,000 flex) — $2.00 of token spend per dollar of seat, the best rate on the ladder. |
| Business | $19/seat | — | 1,900 AI credits per seat, pooled at the billing entity — exactly $1.00 of token spend per dollar of seat, against $1.50 on Pro and $2.00 on Max. |
| Enterprise | $39/seat | — | 3,900 AI credits per seat, pooled at the billing entity — exactly $1.00 of token spend per dollar of seat, against $1.50 on Pro and $2.00 on Max. |
 windsurf.com/pricing now redirects to Cognition's combined Devin and Windsurf pricing page. | |||
| Free | $0 | — | |
| Pro | $20 | — | |
| Max | $200 | — | |
| Teams | $80 | — | $80/month for the team plus $40/month per full developer seat. |
 Companies with more than $1M in annual gross revenue must use Pro or Mega. Midjourney's plan-comparison page blocks automated fetches (HTTP 403); the 9 October 2026 re-check confirmed $10/$30/$60/$120 monthly and $8/$24/$48/$96 annual against secondary pricing trackers rather than the vendor page. | |||
| Basic | $10 | $8 | |
| Standard | $30 | $24 | |
| Pro | $60 | $48 | |
| Mega | $120 | $96 | |
 Annual billing bills 10 months for 12. Credits shown on the pricing page: 10k Free, 30k Starter, 121k Creator, 600k Pro, 1.8M Scale, 6M Business. | |||
| Free | $0 | — | |
| Starter | $6 | $5 | |
| Creator | $22 | $18.33 | First month 50% off. |
| Pro | $99 | $82.50 | |
| Scale | $299 | $249.17 | |
| Business | $990 | $825 | |
 | |||
| Free | $0 | — | |
| Standard | $15 | $12 | |
| Pro | $35 | $28 | |
| Max | $95 | $76 | |
 Annual billing is offered at a discount; Suno's published yearly figures do not reconcile with its own stated savings percentage, so only monthly rates are tracked here. | |||
| Free | $0 | — | 50 credits/day; no song downloads. |
| Pro | $8 | — | 2,500 credits and 20 downloads/month. |
| Premier | $24 | — | 10,000 credits and 60 downloads/month. |
 US list prices. The annual commitment saves up to 35%, the widest monthly/annual gap among the audio tools tracked here. | |||
| Free | $0 | — | 1 media hour/month. |
| Hobbyist | $24 | $16 | 10 media hours, 400 AI credits. |
| Creator | $35 | $24 | 30 media hours, 800 AI credits. |
| Business | $65 | $50 | 40 media hours, 1,500 AI credits. |
 | |||
| Basic | $0 | — | 300 transcription minutes/month. |
| Pro | $16.99/seat | $8.33/seat | 1,200 in-app recording minutes/month. |
| Business | $30/seat | $19.99/seat | Unlimited meetings and in-app recordings. |
 Entry is a 7-day trial of the full feature set rather than a perpetual free tier. Renamed from Pro/Business; the annual discount is 50%, the steepest tracked here. | |||
| Core | $16/seat | $8/seat | 1 hour/day accent conversion. |
| Advanced | $30/seat | $15/seat | 4 hours/day accent conversion, speaker side. |
Standard rates in USD, checked Oct 9, 2026. Click a column to sort. "Model it" opens the calculator with that model selected.
| Model | Long prompts | Discount | Notes | |||
|---|---|---|---|---|---|---|
| claude-fable-5-1 Anthropic · Claude Fable 5.1 · Model it | $10 | $0.25 | $50 | Flat to 1M | Batch 50% off | |
| claude-opus-5-5 Anthropic · Claude Opus 5.5 · Model it | $4 | $0.20 | $20 | Flat to 1M | Batch 50% off | |
| claude-sonnet-5-5 Anthropic · Claude Sonnet 5.5 · Model it | $2 | $0.10 | $10 | Flat to 1M | Batch 50% off | |
| claude-haiku-5-5 Anthropic · Claude Haiku 5.5 · Model it | $0.10 | $0.010 | $0.50 | ×5 in, ×5 out above 100K | Batch 50% off | |
| claude-haiku-4-5-20251001 Anthropic · Claude Haiku 4.5 · Model it | $1 | $0.10 | $5 | Flat | Batch 50% off | |
| claude-opus-5 Anthropic · Claude Opus 5 · Model it | $5 | $0.50 | $25 | Flat to 1M | Batch 50% off | |
| claude-sonnet-4-6 Anthropic · Claude Sonnet 4.6 · Model it | $3 | $0.30 | $15 | Flat to 1M | Batch 50% off | |
| claude-sonnet-4-5-20250929 Anthropic · Claude Sonnet 4.5 · Model it | $3 | $0.30 | $15 | Flat | Batch 50% off | Retires Nov 30, 2026. Details |
| gpt-6-astra OpenAI · GPT-6 Astra · Model it | $10 | $1 | $50 | ×2 in, ×1.5 out above 272K | Batch 50% off | |
| gpt-6.1-sol OpenAI · GPT-6.1 Sol · Model it | $2 | $0.10 | $10 | ×2 in, ×1.5 out above 272K | Batch 50% off | |
| gpt-6-sol OpenAI · GPT-6 Sol · Model it | $2 | $0.20 | $10 | ×2 in, ×1.5 out above 272K | Batch 50% off | |
| gpt-6-luna OpenAI · GPT-6 Luna · Model it | $0.10 | $0.010 | $0.50 | ×2 in, ×1.5 out above 272K | Batch 50% off | |
| gpt-5.6-sol OpenAI · GPT-5.6 Sol · Model it | $4 | $0.40 | $20 | ×2 in, ×1.5 out above 272K | Batch 50% off | Promotional rate guaranteed through Nov 21, 2026; list price was $5 / $30. |
| gpt-5.6-terra OpenAI · GPT-5.6 Terra · Model it | $2 | $0.20 | $12 | ×2 in, ×1.5 out above 272K | Batch 50% off | |
| gpt-5.6-luna OpenAI · GPT-5.6 Luna · Model it | $0.20 | $0.020 | $1.20 | ×2 in, ×1.5 out above 272K | Batch 50% off | |
| gpt-4.1 OpenAI · GPT-4.1 · Model it | $2 | $0.50 | $8 | Flat | Batch 50% off | |
| gpt-4o OpenAI · GPT-4o · Model it | $2.50 | $1.25 | $10 | Flat | Batch 50% off | |
| gpt-4o-mini OpenAI · GPT-4o mini · Model it | $0.15 | $0.075 | $0.60 | Flat | Batch 50% off | |
| o4-mini OpenAI · Model it | $1.10 | $0.28 | $4.40 | Flat | Batch 50% off | Retires Oct 23, 2026. Details |
| gemini-3.1-pro-preview Google · Gemini 3.1 Pro (preview) · Model it | $2 | $0.20 | $12 | ×2 in, ×1.5 out above 200K | Batch 50% off | Preview model. |
| gemini-3.8-flash Google · Gemini 3.8 Flash · Model it | $0.75 | $0.075 | $3.75 | Flat | Batch 50% off | Introductory until Dec 31, 2026, then $1.50 / $7.50. |
| gemini-3.5-flash Google · Gemini 3.5 Flash · Model it | $1.50 | $0.15 | $9 | Flat | Batch 50% off | |
| gemini-3.5-flash-lite Google · Gemini 3.5 Flash-Lite · Model it | $0.30 | $0.030 | $2.50 | Flat | Batch 50% off | |
| gemini-3.1-flash-lite Google · Gemini 3.1 Flash-Lite · Model it | $0.25 | $0.025 | $1.50 | Flat | Batch 50% off | Retires May 7, 2027. Details |
| gemini-2.5-pro Google · Gemini 2.5 Pro · Model it | $1.25 | $0.13 | $10 | ×2 in, ×1.5 out above 200K | Batch 50% off | |
| gemini-2.5-flash Google · Gemini 2.5 Flash · Model it | $0.30 | $0.030 | $2.50 | Flat | Batch 50% off | |
| grok-4.7 xAI · Grok 4.7 · Model it | $2 | $0.50 | $6 | ×2 in, ×2 out above 200K | None | |
| grok-4.3 xAI · Grok 4.3 · Model it | $1.25 | $0.20 | $2.50 | ×2 in, ×2 out above 200K | Batch 20% off | |
| grok-build-0.1 xAI · Grok Build 0.1 · Model it | $1 | $0.20 | $2 | ×2 in, ×2 out above 200K | None | |
| deepseek-v4-pro DeepSeek · DeepSeek V4 Pro · Model it | $0.66 | $0.022 | $1.98 | Flat | Off-peak is 50% below peak | Shown at off-peak rates. Weekday peak hours (01-04 and 06-10 UTC) cost 2×. |
| deepseek-flash DeepSeek · DeepSeek V4.1 Flash · Model it | $0.15 | $0.003 | $0.60 | Flat | Off-peak is 50% below peak | Shown at off-peak rates. Weekday peak hours (01-04 and 06-10 UTC) cost 2×. |
| mistral-large-4 Mistral · Mistral Large 4 · Model it | $1.36 | $0.14 | $4.18 | Flat | Batch 50% off | |
| mistral-medium-3.5 Mistral · Mistral Medium 3.5 · Model it | $1.50 | — | $7.50 | Flat | None | |
| mistral-small-4 Mistral · Mistral Small 4 · Model it | $0.15 | — | $0.60 | Flat | None |
Not included: cache-write premiums (1.25× input on Anthropic and OpenAI), priority tiers, regional processing surcharges (10% on several platforms), tool fees such as web search, and image, audio or video tokens.
Newest first. Allowance changes are listed too: a smaller included usage at the same price is a price rise by another name.
Google · Gemini app (Free and Google AI Plus)
Free and Plus include Gemini 3.1 Pro access→Free: Flash-Lite only. Plus: no Pro model
The free-tier change takes effect on 9 October 2026. Google AI Plus subscribers lose the Pro model on a date Google is emailing individually rather than publishing. Prices are unchanged.
Read the analysis →OpenAI · GPT-6.1 Sol Ultrafast
No published Ultrafast rate for Sol; Astra Ultrafast only, at $60 in / $300 out→$12 in / $60 out (6x standard), US and EU data residency
OpenAI turned Ultrafast on for GPT-6.1 Sol across the API, Codex and ChatGPT Work at a flat 6x multiplier, and it is the first OpenAI speed tier purchasable under EU data residency — GPT-6 Astra Ultrafast remains global-and-US-only, so a European buyer can buy latency on the mid-tier model and not on the flagship. Standard Sol rates are unchanged at $2/$10.
Read the analysis →Anthropic · Claude Haiku 5.5
Haiku 4.5 at $1 in / $5 out, flat→$0.10 in / $0.50 out below 100K-token prompts; $0.50 in / $2.50 out above
Every rate multiplies by five across a 100,000-token prompt threshold — the steepest long-prompt surcharge any major vendor charges, at the lowest threshold. No other Claude model has a length tier. Because the threshold is counted in Anthropic's newer tokenizer, it arrives at roughly 77,000 older-tokenizer tokens.
Read the analysis →Anthropic · Claude Sonnet 5.5
$0.20 cached input per million tokens→$0.10 cached input per million tokens
Lowered from $0.20 to $0.10 per million tokens on 7 October 2026; cache writes and all other rates unchanged. Sonnet 5.5's launch claim of complete rate-card parity with Sonnet 5 has lapsed in the buyer's favour, and it now matches GPT-6.1 Sol's cached-input rate.
Read the analysis →Mistral · Mistral Large 4
Mistral Medium 3.5 at $1.50 in / $7.50 out, 256K context→$1.36 in / $0.14 cached / $4.18 out list, 1M context
The new flagship is 9% cheaper than Medium 3.5 on input and 44% cheaper on output with four times the context, which leaves the mid-tier with no price argument. A launch sale is charging exactly half ($0.68 / $0.07 / $2.09) with no end date published by Mistral; budget at the list price.
Read the analysis →OpenAI · ChatGPT Pro 500
$500/month, 25x Plus usage, Ultrafast
Launched at DevDay alongside a reopened Pro 200. Per unit of Plus allowance it costs exactly what Plus does.
Read the analysis →OpenAI · GPT-6.1 Sol
GPT-6 Sol: $0.20 cached input→$2 in / $10 out, $0.10 cached input
Same list price as GPT-6 Sol and as Claude Sonnet 5.5, with half the cache-read rate. Prompts over 272K tokens bill at 2x input and 1.5x output.
Read the analysis →Anthropic · Claude Sonnet 5.5
Sonnet 5: $2 in / $10 out→$2 in / $10 out
Identical rate card, five documented breaking changes. Against Sonnet 4.5, which retires on 30 November, the newer tokenizer turns a 33% list cut into roughly 13% for the same text.
Read the analysis →Anthropic · Claude API refusals
Refusals free→Input tokens billed for bio, frontier_llm and reasoning_extraction refusals
Refusals in three of five categories now bill input tokens at the full model rate even when nothing is returned. Subscriptions are unaffected.
Read the analysis →OpenAI · GPT-6 Sol and GPT-6 Luna
GPT-5.6 Sol $4/$20→Sol $2/$10, Luna $0.10/$0.50
A flat 50% below the GPT-5.6 tier, stated as permanent. GPT-6 Astra stayed at $10/$50, so it now costs 5x Sol instead of 2.5x.
Read the analysis →Anthropic · Claude Opus 5.5
Opus 5: $5 in / $25 out, $0.50 cache reads→$4 in / $20 out, $0.20 cache reads
Anthropic's "40% cheaper to run" is half rate card and half a default effort level that dropped from high to medium.
Read the analysis →xAI · Grok 4.7
Grok 4.6: $2 in / $6 out→$2 in / $6 out
Unchanged price on a larger base model trained to work longer, so cost per task can rise on a flat rate card. No batch discount.
Read the analysis →Anthropic · Claude Code (Pro, Max, Team)
Temporary +50% boost→Permanent +25% over the pre-May base
Seat prices unchanged. Anthropic has never published the absolute weekly figure the percentages refer to.
Read the analysis →DeepSeek · DeepSeek V4.1 Flash
$0.15 in / $0.60 out off-peak
Peak hours on weekdays cost double. A planned forced migration from V4 Pro was reversed on 14 September.
Read the analysis →OpenAI · GPT-6 Astra
$10 in / $50 out
OpenAI argued price per task, not per token. Prompts over 272K tokens bill at 2x input and 1.5x output for the whole request.
Read the analysis →Google · Gemini 3.8 Flash
$0.75 in / $3.75 out until 31 Dec
Same price and same expiry as 3.7 Flash three weeks earlier.
Read the analysis →Microsoft · Foundry EU Data Zone deployments
EU Data Zone at 10% over Global→EU Data Zone at 20% over Global; US Data Zone unchanged at 10%
Microsoft's Foundry model deployment pricing update took effect on this date, raising EU Data Zone and non-US Regional prices and launching an APAC Data Zone. Microsoft's own GPT-6 Foundry announcement now prices Data Zone deployments at 10% over Global for the US and 20% for the EU, so residency inside Europe costs double what residency inside the US costs. Azure's public pricing page still renders the GPT-6 rows as '$-' and refers readers to the blog, so the premium is not currently readable from the price list.
Read the analysis →Anthropic · Claude Fable 5.1
$1.00 cache reads→$0.25 cache reads ($10/$50 base unchanged)
Worth close to 50% on cache-heavy coding agents and under 5% on conversational traffic.
Read the analysis →GitHub · Copilot Business and Enterprise
Business 3,000 / Enterprise 7,000 credits per user→Business 1,900 / Enterprise 3,900
Seat prices unchanged; the included budget fell by up to 44%.
Read the analysis →OpenAI · GPT-5.6 Sol
$5 in / $30 out→$4 in / $20 out
The first time an OpenAI frontier tier undercut Claude Opus on both input and output.
Read the analysis →DeepSeek · DeepSeek V4 Pro
$0.435 in / $0.87 out→Today: $0.66 in / $1.98 out off-peak, double at peak
It reversed the usual pattern: a Chinese lab raising prices while US labs were cutting.
Read the analysis →Google · Gemini 3.7 Flash
3.6 Flash: $1.50 in / $7.50 out→$0.75 in / $3.75 out (introductory)
The 3.6 Flash rate was later aligned to the same introductory price and the same 31 December expiry.
Read the analysis →Anthropic · Claude Opus 5
Fable 5: $10 in / $50 out→$5 in / $25 out
Near-Fable quality at half the price; it became the default on Claude Max.
Read the analysis →Anthropic · Claude Sonnet 5
Sonnet 4.6: $3 in / $15 out→$2 in / $10 out
Launched as introductory pricing and made permanent on 11 August 2026. It also introduced the newer tokenizer to the Sonnet line.
Read the analysis →GitHub · GitHub Copilot
Flat monthly plans→Same prices, paid as AI Credits
The monthly fee now buys an equal dollar amount of credits. Completions stay free; agentic users reported 10-50x bill spikes.
Read the analysis →Google · Google AI Ultra
$249.99/month→$99.99 (5x) or $199.99 (20x)
Announced at Google I/O 2026 alongside Gemini 3.5 Flash.
Read the analysis →DeepSeek · DeepSeek V4 Pro
$1.74 in / $3.48 out→$0.435 in / $0.87 out
Announced as a promotion and made permanent in late May; reversed upward in August.
Read the analysis →Google · Google AI Pro
2 TB across Gmail, Drive and Photos at $19.99/month→5 TB across Gmail, Drive and Photos at $19.99/month
Announced 1 April 2026 and rolled out globally with no price change; confirmed still in force on the Google One and Gemini subscription pages on 7 October 2026.
Read the analysis →Every price on this page is read from the vendor's own pricing page or help centre, never from a reseller or a cached comparison site, and each block shows the date it was last verified. Pick Right's automated editorial desk re-checks the oldest entries on every run and logs anything that moved, with a link to the analysis of what the change means for a buyer.
Where a vendor publishes an expiry, as Google does for its introductory Flash pricing, the change goes in Already scheduled the day it is found, so the number to budget for is visible before it lands. Found a price that is out of date? Report it and the fix goes in the corrections log.
ChatGPT has a free tier, then Plus at $20 a month. The Pro plan comes in three sizes: Pro 100 at $100, Pro 200 at $200 and Pro 500 at $500 a month, billed monthly only. Every Pro tier now works out to the same $20 per 1x of the Plus usage allowance; Pro 200's included usage drops from 20x to 10x on 30 October 2026. Prices verified October 8, 2026 against OpenAI's help centre.
Claude Pro is $20 a month billed monthly, or $17 a month on the annual plan ($200 up front). Max costs $100 a month for 5x Pro's usage or $200 for 20x. Team seats are $25 (Standard) or $125 (Premium) a month billed monthly. Prices verified October 9, 2026 against claude.com/pricing.
Of the coding tools tracked here, GitHub Copilot Pro is the cheapest paid plan at $10 a month. Cursor Pro and Windsurf Pro cost $20, and Claude Code comes with Claude Pro at $20. A low sticker price is not the whole bill: Copilot draws agent work from a monthly credit allowance, and every tool here caps heavy use, so compare the included usage as well as the price.
Mostly down on the rate card, and less clearly down on the bill. In the last six months this tracker logged 7 price cuts against 2 price rises, but also 2 cuts to included usage at unchanged prices, and several cuts that touch only one line item such as cached input. 4 changes are already scheduled, including Google doubling every Gemini Flash API rate on 1 January 2027.
Vendors count text in their own tokens. Anthropic's newer models, for example, count roughly 30% more tokens for the same text than its older ones, so a lower per-token price can still be a higher bill. Prices also change with caching, batch processing, long prompts (OpenAI bills 2x input above 272K tokens) and time of day (DeepSeek doubles its rate in weekday peak hours). The API cost calculator applies all of these to a workload you describe.
The data behind this page is free to reuse with a link back (CC BY 4.0): subscriptions and change log, API rates.