frida API

ai-forever/FRIDA · Russian embeddings · 512 ctx · ⚠ needs a prefix at the start of each text: queries "search_query: …", documents "search_document: …" (similarity and dedup "paraphrase: …", classification "categorize: …"). Without a prefix retrieval quality drops several times

Type: embeddings · embeddings · RU-first · 1536-dim (ai-forever/FRIDA)
Context: 512
Access: all plans
Endpoint: /v1/embeddings

Runs on our own GPUs in Russia.

How to connect: change one line — base_url. Get an sk- key in the dashboard.

curl https://api.neuraldeep.ru/v1/embeddings \
  -H "Authorization: Bearer $NEURALDEEP_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"frida","input":"search_query: text"}'

https://neuraldeep.tech/en/models/frida