Voyage 3

Cost & access

Voyage 3 is priced per million embedding tokens (or per request), not like chat completion. Compare dimensions and batch APIs on Voyage AI's pricing page. With a 128K tokens context window, long PDFs or chat histories increase input tokens quickly—trim history or summarize older turns in production.

Functional understanding

• Retrieval embedding model for RAG.
• Modalities: text · License: Proprietary · Released 2024-09-01.
• Best-fit workflows for this model:
• • Semantic search, deduplication, and RAG retrieval—Voyage 3 outputs vectors, not chat prose.
• • Clustering support tickets, docs, or product catalogs by meaning.

Technical foundation

• Voyage AI reports Undisclosed parameters; training data: Undisclosed.
• Context: 128K tokens. Open weights: no.
• Voyage 3 is positioned as a embedding model in the Voyage AI lineup.

First API call

Use Voyage AI's embeddings endpoint; pass Voyage 3 as the model name and store vectors in your DB or vector index.

import cohere

co = cohere.Client("YOUR_API_KEY")
response = co.embed(
    texts=["Text to embed for Voyage 3"],
    model="voyage-3",
    input_type="search_document",
)
print(response.embeddings[0][:5], "...")

Important technical topics

• Prompting Voyage 3: be explicit about output format. Weak: "Analyze this." Better: "Return JSON with fields id, total, date for Voyage AI billing data."
• Temperature: use 0–0.3 for extraction and compliance on Voyage 3; 0.7–1.0 for brainstorming.
• Tokens: Voyage 3 bills by tokens (~¾ word each). Undisclosed parameters affect capability; your bill is driven by context length and call volume.
• Context window (128K tokens): everything in one request—system prompt, tools, RAG chunks, and history—must fit. Truncate or summarize when approaching the limit for Voyage 3.

Real enterprise patterns

• Index docs with Voyage 3, then retrieve top-k chunks before calling a chat model.
• Hybrid search: combine keyword (BM25) + embeddings for better recall.
• Version embedding indexes when you change models—dimensions may differ.
• Monitor drift: re-embed when Voyage AI ships a new embedding revision.

Production & security

• Secrets: never commit keys for Voyage 3; use vault + per-environment rotation.
• PII: mask before inference; log redacted prompts only.
• Observability: trace id per request; log model=voyage-3, tokens in/out, latency.
• Rate limits: handle Voyage AI 429/5xx with exponential backoff and circuit breakers.
• Guardrails: schema-validate JSON; block disallowed topics; cross-check numbers against source docs.

Mini projects with this model

• Semantic FAQ: embed help-center articles with Voyage 3, answer from nearest neighbors.
• Duplicate ticket detector for support queues.
• Product recommendation by description similarity.
• Eval: NDCG@k on labeled query–doc pairs.

Suggested stack

• Language: Python 3.11+
• Embeddings: Voyage 3 (Voyage AI)
• Vector DB: Pinecone, Chroma, pgvector, or Weaviate
• Orchestration: LangChain or LlamaIndex for chunking
• UI: Streamlit or Next.js for internal tools
• APIs: FastAPI

Learning path

• Python basics
• HTTP/REST and environment variables
• Voyage AI embeddings API for Voyage 3
• Vector database fundamentals
• Chunking strategies
• RAG retrieval evaluation
• Deploy index refresh pipeline