Qwen3-pruned-6L-from-0.6B-int8-ov

Description

This is a pruned model, originating from the Qwen/Qwen3-0.6B. The model was built to accompany Qwen/Qwen3-8B and to be used as a draft model in the context of Speculative Decoding.

The pruning was performed by applying the findings from recent layer-wise pruning research, see one of the relevant publications, followed by the accuracy recovery fine-tuning over synthetic data generated by the target model Qwen/Qwen3-8B.

Qwen3-pruned-6L-from-0.6B-int8-ov is a model converted to the OpenVINO™ IR (Intermediate Representation) format with weights compressed to int8 by NNCF.

Quantization Parameters

Weight compression was performed using nncf.compress_weights with the following parameters:

• mode: INT8_ASYM

For more information on quantization, check the OpenVINO model optimization guide.

Compatibility

The provided OpenVINO™ IR model is compatible with:

• OpenVINO version 2025.2 and higher
• Optimum Intel 1.25.3 and higher

Running Model Inference with OpenVINO GenAI

1. Install packages required for using OpenVINO GenAI with Speculative decoding:

pip install -U "openvino-genai>=2025.2.0" huggingface_hub

2. Download models from HuggingFace Hub

import huggingface_hub as hf_hub
 
main_model_id = "OpenVINO/Qwen3-8B-int4-ov"
draft_model_id = "OpenVINO/Qwen3-pruned-6L-from-0.6B-int8-ov"
 
main_model_path = "main"
draft_model_path = "draft"
 
hf_hub.snapshot_download(main_model_id, local_dir=main_model_path)
hf_hub.snapshot_download(draft_model_id, local_dir=draft_model_path)

3. Run model inference using the speculative decoding and specify the pipeline parameters:

import openvino_genai
 
prompt = "What is OpenVINO?"
 
config = openvino_genai.GenerationConfig()
config.num_assistant_tokens = 3
config.max_new_tokens = 128
 
def streamer(subword):
    print(subword, end='', flush=True)
    return False
 
main_device = "CPU"
draft_device = "CPU"
 
draft_model = openvino_genai.draft_model(draft_model_path, draft_device)
 
scheduler_config = openvino_genai.SchedulerConfig()
scheduler_config.cache_size = 2


pipe = openvino_genai.LLMPipeline(main_model_path, main_device, scheduler_config=scheduler_config, draft_model=draft_model)
 
pipe.generate(prompt, config, streamer)

More GenAI usage examples can be found in OpenVINO GenAI library docs and samples

Legal Information

The model is distributed under the Intel Research Use License Agreement.

The original model is distributed under Apache License Version 2.0 license. More details can be found in Qwen3-0.6B.

Disclaimer

Intel is committed to respecting human rights and avoiding causing or contributing to adverse impacts on human rights. See Intel’s Global Human Rights Principles. Intel’s products and software are intended only to be used in applications that do not cause or contribute to adverse impacts on human rights.