Qwen3-0.6B INT4

Cloud CPUAWS Graviton G4

A compact decoder-only causal language model for text generation and instruction following, using GPTQ INT4 weight quantization with dynamic INT8 activations to reduce size and memory use while preserving response quality. It supports long-context generation, ChatML-style prompting, optional reasoning-style outputs, and efficient autoregressive decoding through ONNX Runtime GenAI.View on Hugging Face

Accuracyvs base model

MMLU (5-shot)45.18%↓2.14 pp

Tokens / sec

70.4 tok/s

Time to First Token

About this version

Original Model:: Qwen3-0.6B by Alibaba Cloud — Repository
Optimization/conversion:: Arm-Optimized version for execution on Arm-based platforms.
Converted/optimized by:: Arm
License:: The Original Model and the Optimized Model are subject to Apache-2.0.

This repository contains a converted or optimized version of the Original Model (the “Optimized Model”). The Original Model has been converted or optimized as described above for execution on Arm-based platforms.

No retraining or fine-tuning of the Original Model was performed as part of the conversion or optimization. The conversion or optimization was not intended to change the Original Model’s behavior or intended use.

Original Model and Documentation

For information about the Original Model, including its development, training data, intended uses, limitations and other relevant information, please refer to the Original Model repository. Information in that repository was provided by the original developer or other third parties and, unless expressly stated otherwise, has not been independently verified by Arm.

Licenses and Third-Party Terms

Use of the Original Model and the Optimized Model is subject to the applicable licenses, usage restrictions and other terms identified above and in the relevant repositories. Publication of the Optimized Model does not grant any rights beyond those provided under the applicable license terms.

You are responsible for reviewing those terms and ensuring that your use of the Original Model and the Optimized Model is permitted.

About this version

Original Model:: Qwen3-0.6B by Alibaba Cloud — Repository
Optimization/conversion:: Arm-Optimized version for execution on Arm-based platforms.
Converted/optimized by:: Arm
License:: The Original Model and the Optimized Model are subject to Apache-2.0.

Original Model and Documentation

Licenses and Third-Party Terms

You are responsible for reviewing those terms and ensuring that your use of the Original Model and the Optimized Model is permitted.

Metric	Resultsvs baseline	Change
End to end latency p50	1197.60ms3527.37ms	−2329.77 ms
End to end latency p90	1292.14ms3651.42ms	−2359.28 ms
End to end latency p99	1317.93ms3708.18ms	−2390.24 ms
Peak memory	838.22MB2523.21MB	−1684.99 MB
Average memory	831.68MB2496.97MB	−1665.29 MB
Tokens per second	70.40tok/s22.17tok/s	+48.23 tok/s
Time to first token	334.11ms818.64ms	−484.53 ms
MMLU (5-shot)	45.18%47.32%	−2.14%

Configuration	Baseline	Arm Optimized
Dataset	MMLU (test)	MMLU (test)
Sample count	5,000	5,000
Batch size	1	1
Prompt length	127	127
Generation length	60	60
Deployment mode	N/A	chat
Runs	30	30
Warmup runs	2	2

Qwen3-0.6B INT4

Qwen3-0.6B INT4

Get Started with this Model

Text Generator (ONNX Runtime GenAI)

Model Evaluation

Configuration

Purpose of this Release

Comparability