Methodology
Token charges are computed separately for the initial corpus, monthly refresh, and query embeddings. Raw storage assumes 32-bit floats before database indexes or metadata.
Token and API Costs
Estimate corpus indexing, refresh, query embedding, and approximate vector-storage footprint.
Interactive Tool
| Result | Value | How to read it |
|---|---|---|
| Initial embedding cost | $10.00 | One complete corpus pass. |
| Monthly refresh tokens | 5M | Corpus volume re-embedded each month. |
| Monthly query tokens | 15M | Query embedding demand. |
| Recurring embedding cost | $2.00 | Refresh plus query token charges. |
| Raw vector storage | 6.14 GB | Float32 vectors before database overhead. |
Raw vector storage assumes 32-bit floats and excludes indexes, metadata, replicas, backups, and database overhead.
Catalog model: Anthropic Claude Fable 5 →
Inputs
| Input | How it is used |
|---|---|
| Corpus tokens and chunks | Source volume and resulting vector count. |
| Refresh rate | Share of the corpus re-embedded each month. |
| Query volume | Daily query tokens sent to the embedding model. |
| Dimensions and price | Vector dimensions and input price per million tokens. |
Token charges are computed separately for the initial corpus, monthly refresh, and query embeddings. Raw storage assumes 32-bit floats before database indexes or metadata.
Use the storage number as a floor. Real vector databases add index structures, replicas, metadata, backups, and free-space overhead.
Boundaries
Catalog values retain their source and freshness on the linked model, provider, benchmark, or comparison page. Editable scenario assumptions are not Model Markets measurements.
Continue the Analysis
| Tool | Next question |
|---|---|
| Embedding Model Selector | Narrow embedding candidates for retrieval, clustering, or semantic search without mixing them with chat models. |
| Model Migration | Calculate payback for switching a production workload instead of comparing token prices alone. |
| API Usage Calculator | Translate request-level product assumptions into monthly token demand and spend. |
Questions
Separate one-time corpus embedding cost from recurring refresh and query demand. It returns initial tokens and cost, monthly refresh and query cost, vector count, and raw vector storage estimate.
Where the calculation needs model facts, it uses the current Model Markets catalog snapshot updated 2026-09-02. User-entered assumptions remain clearly editable, and unsupported values stay unknown rather than being inferred.
Tokenization and chunk overlap can materially expand corpus tokens. Dimensionality reduction and quantization change storage. Hosted vector database pricing is excluded. Open the linked canonical records and primary sources before making a production or purchasing decision.