Embedding cost calculator
Enter the size of your documents and compare what embedding them costs on each model, now and each month.
Free, runs in your browser, no sign-up. Updated 2026-10-25.
Result
10,000 pages ≈ 7.3M tokens to embed (with overlap). Cheapest: voyage-4-lite at $0 once and $0.0073 a month.
- Corpus tokens
- 6.65M
- Embedded (with overlap)
- 7.32M
- Re-embedded / month
- 0.37M
- Models compared
- 5
| Model | Per 1M tokens | One-off | Monthly | Price checked |
|---|---|---|---|---|
| voyage-4-liteVoyage AI. First 200M tokens free per account. Batch API is 33% off; free tokens do not apply to batch. | $0.02 | $0 | $0.0073 | 2026-10-25 |
| voyage-4Voyage AI. First 200M tokens free per account. Batch API is 33% off; free tokens do not apply to batch. | $0.06 | $0 | $0.022 | 2026-10-25 |
| voyage-4-largeVoyage AI. First 200M tokens free per account. Batch API is 33% off; free tokens do not apply to batch. | $0.12 | $0 | $0.044 | 2026-10-25 |
| voyage-code-4Voyage AI. Code retrieval model. Batch API is 33% off. | $0.12 | $0 | $0.044 | 2026-10-25 |
| Gemini Embedding 2Google. Text input price, paid tier. Image, audio and video inputs are priced separately. | $0.2 | $1.46 | $0.073 | 2026-10-25 |
Assumptions
Every number the result depends on, and where it comes from. Change any value and the result updates.
| Assumption | Value | Source |
|---|---|---|
| Tokens per word | tokens | OceanLab planning assumption: common rule of thumb for English prose (see the token counter). Edit it for your language or for code.Reviewed 2026-10-25 |
| Words per page | words | OceanLab planning assumption: a dense single-spaced page. Edit it for your documents.Reviewed 2026-10-25 |
| Chunk overlap | % | OceanLab planning assumption: chunks usually share some text with their neighbour so context is not cut mid-thought; overlapping text is embedded twice.Reviewed 2026-10-25 |
| Share of the corpus re-embedded each month | % | OceanLab planning assumption: new and edited documents. Set 0 for a static corpus.Reviewed 2026-10-25 |
Ask OceanLab about your RAG build
Opens an email to OceanLab with this result and a link to your inputs. Nothing is sent until you press send, and there is no sign-up.
How this is calculated
Tokens = pages × words per page × tokens per word (or the words or tokens you enter). Chunk overlap is added, because overlapping text is embedded twice.
One-off cost = embedded tokens × the model's price per 1M tokens, after any free allowance. Monthly cost re-embeds the share of the corpus that changes.
Only models with a price read on the vendor's own pricing page are listed. OpenAI, Cohere and Mistral embeddings are left out because no price was found on their pricing pages on the checked date.
Sources
- Google embedding pricing · checked 2026-10-25
- Voyage AI embedding pricing · checked 2026-10-25
Questions about this tool
Which embedding models are included?
Only models whose price was read on the vendor's official page on the checked date. Models without a published price there are left out rather than guessed.
Why add chunk overlap?
Chunks usually share some text with their neighbour, and that text is embedded twice. The overlap share is an editable assumption.
Are free tiers included?
Optionally. A free allowance is applied to the first embedding run, not to batch jobs or monthly re-embedding.
Where do the prices come from?
Each price is read from the provider's official pricing page and shows the date it was checked. Prices that could not be verified are left out, or start at 0 for you to fill in.
Other free tools
OceanLab product builds like this
Products OceanLab designed and built. Each case study covers the screens, the scope and the rules it was built for.
Kaganek - RAG Knowledge Assistant with Citations and Permissions Intact
Designed and built a clickable internal knowledge assistant for a Kraków consulting engineering firm: staff ask in Polish or English and get answers built only from the firm's own SharePoint, Confluence, Google Drive and file share documents, with every claim linked to its source passage and an honest "I don't know" when the sources are weak.
Lukturis - Permission-Aware Enterprise Search Across Seven Internal Tools
Designed and built a clickable enterprise search product for a 600-person Riga software company: one search box over Google Drive, Notion, Jira, Slack, GitHub, HubSpot and the HR policy site, with every result filtered by what the person can already open in the source tool, extractive answer cards that always link to the source, review nudges for content owners and an admin view of connector health and index freshness.
Chaffline - E-Discovery Search and Document Review Platform
Designed and built a clickable e-discovery platform for a Chicago litigation support firm that hosts and reviews data for US law firms: a dashboard across every matter, each matter's path from collection to production, Boolean, proximity and concept search, and a review pane where contract reviewers code documents with the privilege basis recorded at the moment of the call.
