Skip to content

Qwen

Qwen/Qwen3-Embedding-0.6B

Dense text embedding recipe for Qwen3 last-token embeddings.

Start this recipeView on Hugging Face
Model metadata
CategoryEmbeddingParams0.6BContext32,768BackendsCUDA, METALTasksEmbedding
Recipe authorship · Recorded configuration verification

Recipe authorship: Curated

Curated and generated describe how this recipe was written, not whether it ran successfully.

Recorded configuration verification

No current matching configuration verification record.

Records cover the tested default configuration on the named hardware. Matching release minor, variant, and rendering hash are required; changing options does not verify the new configuration.

Choose your actual device. Your browser cannot detect GPU memory.

Command builder

Hardware
Variant

Qwen/Qwen3-Embedding-0.6B

Mode: Server
Advanced
Maximum input length

Context length is measured in tokens. Use the slider or enter a number.

Embedding dimensions
Features

Estimated memory use

approx

Approximate memory allocation in GB. These estimates are not measured peak usage.

  • Weights: ≈ 1.1 GB
  • KV: ≈ 0.0 GB
  • Drafter: ≈ 0.0 GB
  • Runtime allowance: ≈ 1.0 GB

Estimated to leave ≈ 8.6 GB of the ≈ 10.7 GB available memory.

Next: prepare the executable

Notes

  • bf16 weights run as f16 on Apple Silicon.
  • This tier is estimated from nearby measurements.

1. Install

Install in your terminal. Uncached model IDs download automatically on first use.

MLXCEL_INSTALL_DIR="${MLXCEL_INSTALL_DIR:-$HOME/.local/share/mlxcel}" \
  && mkdir -p "$MLXCEL_INSTALL_DIR" \
  && curl -fL -o "$MLXCEL_INSTALL_DIR/mlxcel.zip" https://github.com/lablup/mlxcel/releases/download/v0.8.1/mlxcel-macos-aarch64.zip \
  && unzip -o "$MLXCEL_INSTALL_DIR/mlxcel.zip" -d "$MLXCEL_INSTALL_DIR/mlxcel-macos-aarch64" \
  && export PATH="$MLXCEL_INSTALL_DIR/mlxcel-macos-aarch64:$PATH" \
  && mlxcel --version
Installation and download help

Run the install command in your terminal. The selected executable should then be available on PATH. If it is not found, repeat the displayed PATH export in the same terminal.

Model IDs download automatically when first used. The optional download command prepares the checkpoint ahead of time. Uncached models require internet access and enough disk space; gated models also require access approval and credentials.

2. Run

Run this command in the same terminal after preparation. For server modes, keep it running and wait for the listening/ready message. If memory is insufficient, select a smaller checkpoint or reduce context/concurrency in advanced settings.

Copy all includes the selected installation, checkpoint download, and displayed run commands, in that order. Run them in the same shell; multi-node commands belong on their respective machines.

mlxcel serve

This installation provides the full CLI. Use mlxcel serve; select a source build to use the standalone server.

mlxcel serve -m Qwen/Qwen3-Embedding-0.6B --embedding --alias qwen3-embedding-0.6b \
  --embedding-max-length 8192
Command flag help
-m
Selects the model or checkpoint repository.
--embedding
--embedding
--alias
Sets the served model name used by API clients.
--embedding-max-length
Sets the maximum embedding input length.

3. Check the first result

For a server, use a second terminal after it is ready. A text chat request returns JSON with choices[0].message.content; specialized endpoints return their task output. Replace sample file paths with your own inputs where shown. Connection refused: check the server address/port and terminal logs. Model access/download errors: check the model ID, permissions, internet connection, and free disk space.

curl http://127.0.0.1:8080/v1/embeddings \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3-embedding-0.6b","input":["unified memory","discrete GPU"]}'

Performance evidence

No benchmark is recorded for this selected checkpoint and hardware combination. Commands and memory estimates use the catalog's model information.

Memory fit is a model-based estimate, separate from measured throughput and successful inference verification.

No measurement is available for this hardware yet. See the family page for related profiles. Family

Prerequisites

  • Install the mlxcel build for your selected hardware and check the selected checkpoint's memory estimate.
  • Allow internet access and enough disk space to download an uncached checkpoint; obtain Hugging Face access approval when the checkpoint requires it.

Recipe options

  • • This recipe loads the original checkpoint and does not include an MLX Community 4-bit variant.

Family

Qwen3-Embedding (last-token)

Embedding

Sibling recipes

pending

Guide

Overview

Qwen3 Embedding 0.6B converts text into embedding vectors. This recipe loads Qwen/Qwen3-Embedding-0.6B directly; it does not include an MLX Community 4-bit variant.

Prerequisites

Use mlxcel embed to generate vectors locally, or load the checkpoint in the server's embedding worker. To serve a chat model alongside it, configure the chat model separately.

Gotchas

This recipe omits text-generation controls such as reasoning, draft decoding, and KV-cache compression. Start with a low n_parallel value and measure throughput, queue length, and memory use before increasing concurrency. Embedding micro-batches use memory differently from chat generation.

References