vai models
List all available Voyage AI models with architecture details, pricing, dimensions, and benchmark scores.
- CLI
- Playground
Synopsis
vai models [options]
Description
vai models displays the full Voyage AI model catalog, including embedding models, reranking models, multimodal models, and domain-specific models. By default, it shows only current models; use --all to include legacy models.
Options
| Flag | Description | Default |
|---|---|---|
-t, --type <type> | Filter by type: embedding, reranking, or all | all |
-a, --all | Show all models including legacy | — |
-w, --wide | Wide output (all columns untruncated) | — |
-b, --benchmarks | Show RTEB benchmark scores | — |
--json | Machine-readable JSON output | — |
-q, --quiet | Suppress non-essential output | — |
Examples
List all current models
vai models
Show only embedding models with benchmarks
vai models --type embedding --benchmarks
Include legacy models
vai models --all
JSON output for scripting
vai models --json | jq '.[].name'
Viewing Models in the Playground
Model information is integrated throughout vai playground. When selecting a model in any tab (Embed, Search, Similarity, Rerank), you can see the available models with their key attributes.
Where Models Appear
Model dropdowns: Every tab that calls the Voyage AI API includes a model selector. The dropdown shows model names with their type and pricing.
Estimate tab: The cost estimation view lets you compare models side by side, showing per-token pricing and projected costs for your workload.
Model Information
Each model dropdown provides:
- Model name and type (embedding, reranking, multimodal)
- Pricing per million tokens
- Supported dimensions (for embedding models)
- Context window size
For detailed benchmarks and architecture information, use the CLI with vai models --benchmarks or explore topics with vai explain.
Current Models
| Model | Type | Context | Dimensions | Price | Best For |
|---|---|---|---|---|---|
voyage-4-large | Embedding | 32K | 1024* | $0.12/1M | Best quality, MoE |
voyage-4 | Embedding | 32K | 1024* | $0.06/1M | Balanced |
voyage-4-lite | Embedding | 32K | 1024* | $0.02/1M | Lowest cost |
voyage-code-3 | Embedding | 32K | 1024* | $0.18/1M | Code retrieval |
voyage-finance-2 | Embedding | 32K | 1024 | $0.12/1M | Finance |
voyage-law-2 | Embedding | 16K | 1024 | $0.12/1M | Legal |
voyage-multimodal-3.5 | Multimodal | 32K | 1024* | $0.12/M + $0.60/B px | Text + images |
rerank-2.5 | Reranking | 32K | — | $0.05/1M | Best quality reranking |
rerank-2.5-lite | Reranking | 32K | — | $0.02/1M | Fast reranking |
* Supports flexible dimensions: 256, 512, 1024 (default), 2048
Related Commands
vai explain— Learn about model architectures and embedding conceptsvai estimate— Compare model costs for your workloadvai embed— Generate embeddings with a specific model