Model comparison
Compare Voyage Context 4 and Mistral Embed using the same provider-sourced embeddings & vector search rubric. No mystery score and no invented benchmark ranking.
Facts checked September 4, 2026
Set your usage. Your estimate updates as you type.
Assumes 600 tokens per page, processed separately. Actual token counts vary. This covers embedding only, not storage, search, or generated answers.
Estimates exclude taxes, tools, cache storage/writes, free allowances and custom discounts. Image estimates cover output only, not prompt or reference-image charges. Quality modes differ by model. Unlisted settings are not treated as free.
| Model | Access | Estimated total (USD) |
|---|---|---|
| Mistral EmbedMistral AI | Mistral AI | $0.0006 |
| Voyage Context 4Voyage AI | Voyage AI | $0.0007 |
Quick take
Voyage AI's document-aware embedding model that automatically creates vectors for chunks while preserving information from the surrounding document.
Automatic chunking is convenient but less application-controlled than a custom parser. Test headings, tables, citations, metadata, and update behavior before adopting it across a document estate.
Mistral's straightforward hosted text embedding model for semantic search, clustering, classification, and RAG.
The 8K input window and fixed 1,024 dimensions offer fewer controls than newer embedding families. Verify regional endpoint support and measure retrieval quality on your language and domain.
Compare the published facts
Values use each provider's own published units and limits. A blank means the provider did not publish a directly comparable value in the sources reviewed.
| Embeddings & vector search | Voyage Context 4 | Mistral Embed |
|---|---|---|
| Embedding priceCurrent provider price per million input tokens or the closest published billing unit. | $0.12 / 1M tokens after 200M free | $0.10 / 1M tokens |
| Input capacityMaximum content accepted in one embedding input, using the provider's documented token basis. | 32K per internal pass; longer documents split automatically | 8K tokens |
| Vector dimensionsSupported output sizes; smaller vectors reduce storage while larger vectors may preserve more information. | 256, 512, 1,024, or 2,048 | 1,024 |
| Accepted inputsText, code, image, audio, video, PDF, or document-aware input supported by the endpoint. | Whole text documents | Text |
| Retrieval controlsQuery/document modes, task types, truncation, chunking, or other controls that shape vectors for retrieval. | Automatic contextual chunking, overlap, and long-document splitting | Single or batched inputs; application handles chunking |
| Where it runsDirect API, cloud marketplace, private deployment, or self-hosted route documented by the provider. | Voyage hosted API and MongoDB-integrated routes | Mistral hosted API and Studio |
How to choose
Start with the job you need to complete, then validate cost, access, and policy details on your exact provider route.
Voyage's public terms allow customer content to improve services unless the customer opts out. Eligible paid organizations can configure an opt-out and zero-day retention; separately negotiated enterprise terms may differ.
Provider and API links
Mistral says API data is not used for training by default. Standard API inputs and outputs are generally retained for 30 rolling days for abuse monitoring unless approved zero-data-retention controls apply.
Provider and API links
Frequently asked questions
Voyage Context 4: Voyage AI's document-aware embedding model that automatically creates vectors for chunks while preserving information from the surrounding document. Mistral Embed: Mistral's straightforward hosted text embedding model for semantic search, clustering, classification, and RAG.
Consider Voyage Context 4 when your priority is Long-document RAG. Consider Mistral Embed when your priority is Text-only semantic search. Test both with your own data and provider route before committing.
No. This comparison aligns provider-published facts for the Embeddings & vector search category. It does not claim a universal winner or combine incompatible third-party benchmark scores.
