OPEN PRICING DATA • OFFICIAL SOURCES
AI Pricing
Sourcebook
A source-linked reference for public AI subscription, token, cloud hosting, image, and video API rates. Find the billing unit first, then open the matching calculator to model a real workload.
Providers11
Official-source rows42
Last checkedJul 24
HOW TO READ THE DATA
Rates without units are misleading.
One dollar figure can mean per million input tokens, per generated second, per API credit, per token-hour of cache storage, or one account billed monthly. The constraint column records the first billing exception most likely to change a budget. It is not a substitute for the complete provider terms.
AI SUBSCRIPTION
AI subscription rates
| Provider / product | Public rate | Billing unit | Important constraint | Evidence |
|---|---|---|---|---|
| Google Google AI Plus · US consumer | $4.99 monthly | Per personal Google account | Consumer subscription, not Gemini API. Benefits, usage limits, eligibility, and regional pricing vary. | Official ↗ Calculator → |
| Google Google AI Pro · US consumer | $19.99 monthly | Per personal Google account | Consumer subscription, not Gemini API. Current US bundle includes plan-specific storage and benefits. | Official ↗ Calculator → |
| Google Google AI Ultra · US consumer | $99.99 monthly 5× · $199.99 monthly 20× | Per personal Google account | Two current usage tiers. Starts at 20 TB; features and regional availability vary. | Official ↗ Calculator → |
| Canva Free · US annual list price | $0 annual | One person | Free is a one-person plan. Included features and limits can change; optional purchases are excluded. | Official ↗ Calculator → |
| Canva Pro · US annual list price | $144 annual | One person | Pro is an individual subscription. Tax, regional pricing, promotions, and optional AI Pass cost are excluded. | Official ↗ Calculator → |
| Canva Business · US annual list price | $250 annual | Per person | Canva says Business has no seat minimum. Tax, regional pricing, promotions, and optional AI Pass cost are excluded. | Official ↗ Calculator → |
| Notion Business · US list price | $20 monthly | Per paid workspace member | Core Notion AI is included. Enterprise is custom-priced; Free and Plus have limited trial AI responses. | Official ↗ Calculator → |
| Notion Custom Agent monthly credits | $10 per 1,000 credits | Shared workspace credit balance | Available to Business and Enterprise. Usage varies by task; monthly credits reset and unused credits do not roll over. | Official ↗ Calculator → |
| Perplexity Perplexity Pro | $20 monthly · $200 annual | Per consumer account | No recurring monthly Computer-credit allocation. API credits are separate. | Official ↗ Calculator → |
| Perplexity Perplexity Max | $200 monthly · $2,000 annual | Per consumer account | Includes 10,000 monthly Computer credits. Annual billing is available on the web app; API is separate. | Official ↗ Calculator → |
| Perplexity Enterprise Pro | $40 monthly · $400 annual | Per active enterprise seat | Includes 500 monthly Computer credits per seat; first month has a 30-day seat commitment. | Official ↗ Calculator → |
| Perplexity Enterprise Max | $325 monthly · $3,250 annual | Per active enterprise seat | Includes 15,000 monthly Computer credits per seat. Organizations can mix Enterprise Pro and Max seats. | Official ↗ Calculator → |
API TOKENS
API tokens rates
| Provider / product | Public rate | Billing unit | Important constraint | Evidence |
|---|---|---|---|---|
| OpenAI GPT-5.6 Sol | $5 input · $0.50 cached input · $30 output | Per 1M text tokens | Prompts above 272K input tokens use higher long-context rates; cache writes can cost more than cache reads. | Official ↗ Calculator → |
| OpenAI GPT-5.6 Terra | $2.50 input · $0.25 cached input · $15 output | Per 1M text tokens | Token rates exclude tools, storage, audio, images, priority processing, and other metered features. | Official ↗ Calculator → |
| OpenAI GPT-5.6 Luna | $1 input · $0.10 cached input · $6 output | Per 1M text tokens | Token rates exclude tools, storage, audio, images, priority processing, and other metered features. | Official ↗ Calculator → |
| Google Gemini 3.6 Flash | Standard $1.50 / $7.50 · Batch $0.75 / $3.75 | Input / output per 1M tokens | Cached input is $0.15 Standard or $0.075 Batch. Output includes thinking tokens. | Official ↗ Calculator → |
| Google Gemini 3.5 Flash | Standard $1.50 / $9 · Batch $0.75 / $4.50 | Input / output per 1M tokens | Cached input is $0.15 Standard or $0.075 Batch. Output includes thinking tokens. | Official ↗ Calculator → |
| Google Gemini 3.5 Flash-Lite | Standard $0.30 / $2.50 · Batch $0.15 / $1.25 | Input / output per 1M tokens | Cached input and Search grounding are separate billing components. | Official ↗ Calculator → |
| Google Gemini 3.1 Pro Preview | ≤200K $2 / $12 · >200K $4 / $18 | Input / output per 1M tokens | Preview model. The higher band activates when a prompt exceeds 200,000 input tokens. | Official ↗ Calculator → |
| Anthropic Claude Opus 4.8 | Standard $5 / $25 · Batch $2.50 / $12.50 | Input / output per 1M tokens | Prompt-cache writes are $6.25 for 5 minutes or $10 for 1 hour; reads are $0.50. Fast mode uses separate rates. | Official ↗ Calculator → |
| Anthropic Claude Sonnet 5 | Through Aug 31 $2 / $10 · from Sep 1 $3 / $15 | Input / output per 1M tokens | Introductory pricing ends August 31, 2026. Prompt-cache, Batch, and US-only inference modifiers can stack. | Official ↗ Calculator → |
| Anthropic Claude Sonnet 4.6 | Standard $3 / $15 · Batch $1.50 / $7.50 | Input / output per 1M tokens | Prompt-cache writes are $3.75 for 5 minutes or $6 for 1 hour; reads are $0.30. | Official ↗ Calculator → |
| Anthropic Claude Haiku 4.5 | Standard $1 / $5 · Batch $0.50 / $2.50 | Input / output per 1M tokens | Prompt-cache writes are $1.25 for 5 minutes or $2 for 1 hour; reads are $0.10. | Official ↗ Calculator → |
CLOUD HOSTING
Cloud hosting rates
| Provider / product | Public rate | Billing unit | Important constraint | Evidence |
|---|---|---|---|---|
| Vercel Pro platform | $20 monthly | One deploying seat plus platform access | Includes one deploying seat and $20 of monthly managed-infrastructure usage credit. Unused credit resets each month. | Official ↗ Calculator → |
| Vercel Pro transfer and request allocations · low-cost US regions | 1 TB FDT + 10M Edge Requests included · then $0.15/GB + $2/M | Monthly team allocation, then regional on-demand usage | The low end of Vercel's regional range applies to selected US regions. Other regions can cost more; usage credit applies after allocations. | Official ↗ Calculator → |
| Vercel Fluid Compute · Cleveland, IAD, and PDX | $0.128 CPU-hour · $0.0106/GB-hour · $0.60/M invocations | Active CPU, provisioned memory, and function invocations | CPU pauses during I/O waits, while memory is billed for the instance lifetime handling requests. The first 1M invocations are included; other regions can cost more. | Official ↗ Calculator → |
| Vercel Image Optimization · default pricing | 5K transforms + 300K reads + 100K writes included · then $0.05/1K + $0.40/M + $4/M | Monthly included usage, then regional on-demand usage | Transformations are billed on cache MISS or STALE; cache reads and writes use their own units. Regional rates can be higher. | Official ↗ Calculator → |
IMAGE SUBSCRIPTION
Image subscription rates
| Provider / product | Public rate | Billing unit | Important constraint | Evidence |
|---|---|---|---|---|
| Midjourney Basic plan | $10 monthly · $96 annual | Per account subscription | Includes about 200 Fast GPU minutes; no Relax image mode. Annual price is paid up front. | Official ↗ Calculator → |
| Midjourney Standard plan | $30 monthly · $288 annual | Per account subscription | Includes about 900 Fast GPU minutes and Relax images. Annual price is paid up front. | Official ↗ Calculator → |
| Midjourney Pro plan | $60 monthly · $576 annual | Per account subscription | Includes about 1,800 Fast GPU minutes, Relax images and SD video, and Stealth mode. | Official ↗ Calculator → |
| Midjourney Mega plan | $120 monthly · $1,152 annual | Per account subscription | Includes about 3,600 Fast GPU minutes, Relax images and SD video, and Stealth mode. | Official ↗ Calculator → |
| Midjourney Extra Fast Time | $4 | Per Fast GPU hour | Requires an active subscription and purchased time expires under the provider's current rules. | Official ↗ Calculator → |
VIDEO API
Video API rates
| Provider / product | Public rate | Billing unit | Important constraint | Evidence |
|---|---|---|---|---|
| Runway API credit | $0.01 | Per API credit | Credits per second and minimum credits vary by model. Consumer web-app subscriptions are separate. | Official ↗ Calculator → |
| Runway Gen-4.5 | 12 credits · $0.12 | Per generated second | Every requested generation, retry, and alternative can add billable generated seconds. | Official ↗ Calculator → |
| Runway Gen-4 Turbo | 5 credits · $0.05 | Per generated second | Every requested generation, retry, and alternative can add billable generated seconds. | Official ↗ Calculator → |
| Google Cloud Veo 3.1 · 720p / 1080p · video only | $0.20 | Per generated second | The 4K video-only rate is $0.40 per second. Audio and other sales routes can use different SKUs. | Official ↗ Calculator → |
| Google Cloud Veo 3.1 Fast · 720p / 1080p | $0.08 / $0.10 | Per generated second | The 4K Fast rate is $0.25 per second. Fast and Standard are different model SKUs. | Official ↗ Calculator → |
| HeyGen Avatar V · Digital Twin · 720p / 1080p | $0.0667 | Per generated second | Self-serve API uses a prepaid wallet; a first PAYG purchase can have a funding minimum. | Official ↗ Calculator → |
| HeyGen Avatar IV · Photo Avatar · 720p / 1080p | $0.05 | Per generated second | Rate differs by avatar type. Current public self-serve documentation does not list a 4K row. | Official ↗ Calculator → |
| HeyGen Avatar IV · Digital Twin / Studio · 720p / 1080p | $0.0667 | Per generated second | Rate differs by avatar type. Current public self-serve documentation does not list a 4K row. | Official ↗ Calculator → |
| HeyGen Avatar III · Photo Avatar · 720p / 1080p | $0.0433 | Per generated second | Legacy model availability can change; confirm the current API model list before production. | Official ↗ Calculator → |
| HeyGen Avatar III · Digital Twin / Studio · 720p / 1080p | $0.0167 | Per generated second | Legacy model availability can change; confirm the current API model list before production. | Official ↗ Calculator → |
Dataset checked July 24, 2026. Independent reference; not affiliated with the listed providers. Public prices can change by product version, billing account, contract, region, currency, tax, and later documentation updates. Verify the linked official source before purchasing or launching a workload.