Qwen3-Embedding-4B (Qwen) specs, prices, where it runs and its benchmark standing

Name: Qwen3-Embedding-4B (Qwen)
Author: Alibaba

open weights · top 10 on 1 of 1 · released Jun 5, 2025

Provider: Alibaba
Release date: Jun 5, 2025 · Pricing lists Oct 28, 2025; Boards
Parameters: 4B total; Hugging Face
Modalities: Text to Embeddings; Hugging Face
License: Apache 2.0; Hugging Face
Context window: 33K tokens; Pricing
Pricing: $0.02 in / $0.0 out per 1M tokens; All prices
Hosting: 1 priced API route · self-host: weights on Hugging Face; Pricing
Hardware: Estimate: about 8 GB of weights as published (BF16), about 4 GB at 8-bit, about 2 GB at 4-bit, plus memory for the context; Hugging Face config
Weights: Open weights
Benchmarks: 1 board captured

Rank	Board	Category	Score	Of	Over time
6	MTEBEmbeddings · as “Qwen/Qwen3-Embedding-4B”	Embeddings	69.5%	96

Versions and release history

3 versions in this line since Jun 5, 2025, from Pricing, OpenRouter and the boards.

Version	Released	Source
Qwen3-Embedding-0.6B (Qwen)	Jun 5, 20251 year ago	Boards
Qwen3-Embedding-4B (Qwen)this model	Jun 5, 20251 year ago	Boards
Qwen3-Embedding-8B (Qwen)	Jun 5, 20251 year ago	Boards

Category ranks, more models and catalog changes