Voyage AI / voyageai/voyage-multimodal-3.5

Voyage Multimodal 3.5 - access through LLMTR

Voyage Multimodal 3.5 places text and images in the same vector space, so you can search images with text or find similar images from an image. On pages containing screenshots, tables, charts and diagrams it retrieves directly, without converting the content to text first. A single input may mix text and image parts. Billing has two components: text is charged per token and images per pixel, with images below 50,000 pixels upscaled to 50,000 and those above 2 million downsampled to 2 million.

Technical specifications

Canonical IDvoyageai/voyage-multimodal-3.5
ProviderVoyage AI
Context window32,000 tokens
OperationsEMBEDDINGS
Modalitiesimage, text, embedding

Pricing

An 8% platform margin applies to credit top-ups; model usage prices are not separately marked up.

OperationMetricUnitPrice
EMBEDDINGSINPUT_TEXTPER_1M_TOKENS$0.120000
EMBEDDINGSINPUT_IMAGEPER_1B_PIXELS$0.600000

Example usage

With existing OpenAI SDK flows, change only the base URL and model identifier.

curl https://llmtr.com/v1/chat/completions \
  -H "Authorization: Bearer llmtr-your_key" \
  -H "Content-Type: application/json" \
  -d '{"model":"voyageai/voyage-multimodal-3.5","messages":[{"role":"user","content":"Hello"}]}'

Related models