All models
Qwen
Qwen3 Embedding 4B
Qwen3's economical embeddings
Text
Vector 2560
- Type
- Embeddings
- Context
- 33K tokens
- Russian
- high
- Price
- 1,6 ₽ / 1M tokens
- Vector size
- 2560
- Release date
- June 2025
chat/completions
import OpenAI from "openai"const client = new OpenAI({ baseURL: "https://api.yenisei.ru/v1", apiKey: process.env.YENISEI_API_KEY,})const stream = await client.chat.completions.create({ model: "qwen3-embedding-4b", messages: [{ role: "user", content: "Hi!" }], stream: true,})for await (const chunk of stream) { process.stdout.write(chunk.choices[0]?.delta?.content ?? "")}About the model
Qwen3 Embedding 4B is the lighter embeddings model in the Qwen lineup, balancing quality and cost. It turns text into vectors for semantic search and RAG and keeps good search accuracy at a smaller size and price. The model is multilingual and handles Russian confidently, which is convenient for international products. The resulting vectors work well for clustering, recommendations, deduplication and classification. Thanks to its compact size, it is economical at high volume.
Strengths
- Good search accuracy at a lower price
- Many languages, including Russian
- Economical at high volume
- Vectors for clustering and recommendations
- Suited to RAG and semantic search
Where to use it
- Semantic search across documents
- RAG pipelines and question-answering systems
- Clustering and topic grouping
- Recommendation systems
- Deduplication and similarity search
- Vector-based text classification