qwen3-embedding-8b
Qwen embedding model for semantic search, retrieval, and RAG pipelines.
qwen
40K context
Context rank
#3 of 13
40K token window
Benchmarks
Embedding quality on MTEB, the Massive Text Embedding Benchmark, as published on the Qwen3-Embedding-8B model card. It ranked No.1 on the MTEB multilingual leaderboard at release (5 June 2025).
MTEB mean task score
Best for
Where this model earns its keep.
Semantic search and retrieval
RAG pipelines
Multilingual and code search
The numbers
Pricing is live from our platform. Prices per 1M tokens, zero data retention on every request.
| Input price | $0.02 |
| Embedding dimensions | Up to 4,096 (adjustable from 32) |
| MTEB multilingual / English v2 | 70.6 / 75.2 |
| Context window | 40K tokens |
Quick start
OpenAI-compatible. Switch in one line.
# pip install openai
client = OpenAI(base_url="https://api.tensorx.ai/v1", api_key="tsx-...")
r = client.embeddings.create(
model="qwen/qwen3-embedding-8b",
input=["EU-sovereign inference", "Zero data retention"],
)
# r.data[0].embedding
client = OpenAI(base_url="https://api.tensorx.ai/v1", api_key="tsx-...")
r = client.embeddings.create(
model="qwen/qwen3-embedding-8b",
input=["EU-sovereign inference", "Zero data retention"],
)
# r.data[0].embedding
Benchmark data from MTEB, as published by the model developer. Pricing live from the TensorX platform. All inference on EU-sovereign infrastructure with zero data retention.