LAzy Note Qwen3 Embedding 0.6B Q4_K_M

This repository contains the Q4_K_M GGUF used by LAzy Note's optional on-device Idea and Journal classifier.

Provenance

Upstream: Qwen/Qwen3-Embedding-0.6B-GGUF
Upstream revision: 370f27d7550e0def9b39c1f16d3fbaa13aa67728
Quantization source: Qwen3-Embedding-0.6B-f16.gguf
llama.cpp commit: e3546c7948e3af463d0b401e6421d5a4c2faf565
Quantization: Q4_K_M
Output SHA-256: 44f67b5875def0e90ae67247b141a5a861538efb8d8b0f783eef3d94211d9dcd
Output size: 396474560 bytes

The model was quantized with the pinned llama.cpp llama-quantize binary. No model weights were fine-tuned by LAzy Note.

Application use

LAzy Note downloads this file only after explicit user confirmation. Text is embedded and classified on the device; note text and embeddings are not sent to this repository or another inference service.

The app verifies both the exact file size and SHA-256 before installing the model. The download URL is pinned to a full repository commit.

License

The upstream Qwen3 Embedding model is distributed under Apache License 2.0. See Qwen/Qwen3-Embedding-0.6B-GGUF and NOTICE.md for attribution and derivative details.

Downloads last month: 21

GGUF

Model size

0.6B params

Architecture

qwen3

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for unsig46/lazynote-qwen3-embedding

Base model

Qwen/Qwen3-0.6B-Base

Finetuned

Qwen/Qwen3-Embedding-0.6B

Quantized

(279)

this model