checklist.day

Qwen3 Embedding 4B W4A16 G128

JSON →
boboliu embedding
text

A 4-bit quantized embedding model based on Qwen3 with 4B parameters, using W4A16 and group size 128 for efficient retrieval.

Specs

context window 41K tokens
max output 41K tokens

Capabilities

fine-tunable

Dates

releasedApr 2025

Resources

homepagehuggingface.co/boboliu/Qwen3-Embedding-4B-W4A16-G128 ↗

API

full doc /v1/models/boboliu/Qwen3-Embedding-4B-W4A16-G128