Zen Embedding 0.6B GGUF

0.6B Zen embedding model, quantized to GGUF (Q8_0) for CPU and mixed CPU/GPU inference with llama.cpp and compatible runtimes.

Repackaged from Qwen/Qwen3-Embedding-0.6B (apache-2.0, Alibaba Qwen), quantized to GGUF. Not trained from scratch — a permissively-licensed redistribution for the OSS-clean Zen model line.

Files

Usage

from llama_cpp import Llama

llm = Llama.from_pretrained(
    repo_id="zenlm/zen-embedding-0.6B-GGUF",
    filename="*Q8_0.gguf",
    embedding=True,
)
print(llm.create_embedding("Hello!"))

Full-precision safetensors: zenlm/zen-embedding-0.6B.

License

apache-2.0. Upstream: Qwen/Qwen3-Embedding-0.6B by Alibaba Qwen. This repository redistributes a quantized derivative under the same license.