Open data: Embeddings
Models for semantic search, RAG and code retrieval.
Published values, excluding taxes. A missing value is not zero. Plans may share resources or charge by usage; check their conditions. Third-party indexes are excluded from these downloads.
Most recent review:
Embeddings
| Product | Text price (USD / M tokens) | Reference dimensions (dimensiones) | Input per document (tokens) | Product review date |
|---|---|---|---|---|
| Mistral Embed | 0.1 | 1,024 | 8,000 | |
| Codestral Embed | 0.15 | 3,072 | 8,000 | |
| Voyage 4 Lite | 0.02 | 1,024 | 32,000 | |
| Cohere Embed 5 Pro | 0.12 | 2,048 | 128,000 | |
| Cohere Embed 5 Fast | 0.08 | 2,048 | 128,000 | |
| Voyage 4 | 0.06 | 1,024 | 32,000 | |
| Voyage 4 Large | 0.12 | 1,024 | 32,000 | |
| Cohere Embed 4 | — | 1,536 | 128,000 | |
| Voyage Code 4 | 0.12 | 1,024 | 32,000 |
What each field means
input· USD / M tokens- Embeddings API price for text; index storage is not included.
dimensions· dimensiones- Reference setting listed on the product page. More dimensions mean more storage, not guaranteed higher quality.
context· tokens- Published limit per individual text. A batch limit may differ.
Conditions and sources
Mistral Embed
Text model; vectors aren't compatible with another model's just because they share a dimension. Context published as 8k.
Codestral Embed
3072 is the configurable maximum, not a quality score. A smaller dimension requires regenerating and versioning the index. Context published as 8k.
Voyage 4 Lite
Rate after initial credits; the comparison doesn't deduct credits. Shows the default 1024 dimensions; 256, 512 and 2048 are also supported.
Cohere Embed 5 Pro
Text rate; images cost $0.40 per million tokens. The default 2048 dimensions are shown; it also supports 256, 512, 768, 1024 and 1536, in float, int8 and binary formats.
Cohere Embed 5 Fast
Text rate; images cost $0.40 per million tokens. The default 2048 dimensions are shown; it also supports 256, 512, 768, 1024 and 1536, in float, int8 and binary formats.
Voyage 4
Standard per-token rates; the first 200M tokens have a per-account allowance under provider conditions. Configurable dimensions: the displayed value is the default. Context per input: 32,000 tokens; this is not the batch token limit.
Voyage 4 Large
Standard per-token rates; the first 200M tokens have a per-account allowance under provider conditions. Configurable dimensions: the displayed value is the default. Context per input: 32,000 tokens; this is not the batch token limit.
Cohere Embed 4
No comparable per-token rate is included: price depends on modality and deployment. Dedicated per-instance offerings aren't artificially converted to tokens.
Voyage Code 4
Standard rate: $0.12/M tokens; the official table publishes an initial 200M-token allowance per account. Context per input: 32,000 tokens, distinct from the batch limit. Default dimensions: 1,024; supports 256, 512, 1,024 and 2,048.
License and attribution
Credit Codifly and link to this page when reusing the data. CC BY 4.0
Compare optionsGuides to help you decide
Prices read from official sources, worked examples and criteria for choosing. Download the data