Voyage AI Embedding Cost Calculator
Estimate Voyage AI text embedding and rerank cost. Tab A: voyage-4-large / voyage-4 / voyage-4-lite / voyage-context-4 / voyage-code-4 with optional free-tier toggle (200M tokens). Tab B: rerank-3 / rerank-3-lite plus optional N requests × docs helper. Batch API is 33% off and free allotment does not apply to Batch. Multimodal (pixels) noted in FAQ only. Estimates only; official Voyage pricing prevails.
Math runs in your browser. Nothing is uploaded.
Free tokens applied
—
Billable tokens
—
Effective rate
—
Monthly USD
—
Estimates only; official vendor pricing prevails (2026-10-30). Billable = max(0, usage − free); cost = billable÷1M×rate. Batch = standard ×0.67 and free allotment does not apply. Text embed + rerank only; multimodal pixels in FAQ.Official pricing: Voyage AI pricing
OpenAI Embeddings Cost Calculator · Gemini Embedding Cost Calculator · RAG Cost Calculator
More in this group
- Vector Database Cost Comparison Pinecone vs Upstash Vector monthly cost for one workload.
- Pinecone Cost Calculator Pinecone storage + WU + RU vs $50 minimum → monthly cost.
- Upstash Vector Cost Calculator Upstash Vector PAYG vs Fixed $60 + break-even.
- Document Parsing API Cost Comparison LlamaParse vs Reducto vs Unstructured, $ / 1k pages.
- LlamaParse Cost Calculator LlamaParse credits by tier → plan + $ / 1k pages.
- Reducto Cost Calculator Reducto per-endpoint pages → monthly $ + free credit.
- OpenAI Embeddings Cost Calculator Embedding tokens with Batch and RAG ingest scenario.
- Gemini Embedding Cost Calculator Embedding 2 multimodal or embedding-001 text.
How it works
- How does free allotment work?
- voyage-4 family and voyage-context-4 / voyage-code-4 include 200M free tokens per account (standard endpoints). Toggle “apply free tier” to subtract that allotment from monthly tokens before pricing.
- What about Batch 33% off?
- Batch pays 67% of the standard $/1M rate. Free token credits do not apply to Batch API usage — the calculator disables free allotment when Batch is checked.
- How is rerank usage measured?
- Processed tokens = (query tokens × number of documents) + sum of document tokens. Use the optional N requests × docs helper (defaults approximate the docs’ 100-doc / 500-token example).
- Multimodal embeddings?
- Pixels pricing for voyage-multimodal is noted only; this page covers text embed + rerank.
- Are results invoices?
- No. Estimates only; docs.voyageai.com/docs/pricing prevails.