Home / Tools / Embedding cost
Embedding cost calculator
Estimate corpus indexing and per-query embedding spend from tokens and an editable $/1M rate. Compare reference models and optional batch discounts.
RAG cost · Tokenizer · Tokenizer comparison · Embedding cost guide
Common public list price reference. Verify on OpenAI pricing.
Effective rate $0.020/1M. 0% = standard. Set only if your provider publishes a batch discount for embeddings.
Corpus size
Index volume ≈ 5,000,000 tokens.
Refresh + query embeds
0% = one-time index only. 10% amortizes a full re-embed about every 10 months.
- Index once
- $0.1
- Index amortized / month
- $0.01
- Query embed / call
- $0.0000016
- Query embeds / month
- $0.048
- Embed API / month
- $0.058
- Effective $/1M
- $0.020/1M
| Reference preset | $/1M (eff.) | Index once | Embed / month |
|---|---|---|---|
| OpenAI text-embedding-3-small | $0.020/1M | $0.1 | $0.058 |
| OpenAI text-embedding-3-large | $0.130/1M | $0.65 | $0.377 |
| OpenAI text-embedding-ada-002 (legacy) | $0.100/1M | $0.5 | $0.29 |
Compare table uses the same corpus, batch discount, refresh, and query volume. Rates are reference defaults only.
How to use this
- Pick a preset or Custom, then edit $/1M to match your provider.
- Size the corpus with docs × tokens, a total token count, or a pasted sample scaled by document count.
- Set re-embed share and monthly query embeds. Read index vs query lines.
- For full RAG (generation + retrieval), open the RAG cost calculator.
RAG cost calculator · OpenAI embedding pricing · Batch savings · Reindex cost
Related tools
Related guides
Embedding cost calculator: estimate corpus spend · OpenAI embedding pricing: text-embedding-3 planning · Estimate embedding tokens · Batch embedding cost savings · Embedding reindex cost: how often to re-embed
Sources and references
Official documentation used for definitions, counting methods, or rate cards. Always confirm critical budgets on the provider page.
FAQ
- How is embedding API pricing calculated?
- Cost ≈ (tokens embedded / 1,000,000) × price per million tokens. Embeddings bill input text only. There is no separate “output token” line like chat completions.
- Do embeddings charge for output tokens?
- No. You pay to process the text you send. The API returns a vector; that vector is not billed as generated tokens.
- Are these rates from the TokenCalculator LLM catalog?
- No. Presets are editable reference defaults. Always verify on your embed provider’s pricing page or invoice before budgeting.
- Does Batch API reduce embedding cost?
- Some providers discount async batch embedding jobs. Set the batch discount % to match the published offer (often ~50% on OpenAI when available). Confirm current terms before you rely on it.