We lifted 12 functions out of this paper's own repositories and ran 7 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| efficient-ml/qwen3-quantization | canonical | 7 of 12 |
| Function | Status | Where it lives |
|---|---|---|
| find_layers | Ran | efficient-ml/qwen3-quantization/BiLLM/modelutils.py code served (permissive licence) · get_code("a9e7f2cdf016b88b") |
| high_order_residual | Ran | efficient-ml/qwen3-quantization/BiLLM/binary.py code served (permissive licence) · get_code("7ae3b57ecdeec075") |
| normal_quantize | Ran | efficient-ml/qwen3-quantization/BiLLM/binary.py code served (permissive licence) · get_code("dd1edfd05dda7353") |
| part_mean | Ran | efficient-ml/qwen3-quantization/BiLLM/binary.py code served (permissive licence) · get_code("25eb3a441763f260") |
| quantize_activation_per_token_absmax | Ran | efficient-ml/qwen3-quantization/SmoothQuant-for-Qwen3/smoothquant-main/smoothquant/fake_quant.py code served (permissive licence) · get_code("c1cb961b4f5cb75c") |
| quantize_weight_per_channel_absmax | Ran | efficient-ml/qwen3-quantization/SmoothQuant-for-Qwen3/smoothquant-main/smoothquant/fake_quant.py code served (permissive licence) · get_code("770da8497b2ff4ba") |
| quantize_weight_per_tensor_absmax | Ran | efficient-ml/qwen3-quantization/SmoothQuant-for-Qwen3/smoothquant-main/smoothquant/fake_quant.py code served (permissive licence) · get_code("dd3493503cb95888") |
| get_act_scales | Not yet run | efficient-ml/qwen3-quantization/SmoothQuant-for-Qwen3/smoothquant-main/smoothquant/calibration.py code served (permissive licence) · get_code("3fed3338f2ffb08f") |
| get_ptb | Not yet run | efficient-ml/qwen3-quantization/BiLLM/datautils.py code served (permissive licence) · get_code("a7cbf4fc49c71e03") |
| get_static_decoder_layer_scales | Not yet run | efficient-ml/qwen3-quantization/SmoothQuant-for-Qwen3/smoothquant-main/smoothquant/calibration.py code served (permissive licence) · get_code("a5a75a01c343fb3b") |
| get_tokenizer | Not yet run | efficient-ml/qwen3-quantization/BiLLM/datautils.py code served (permissive licence) · get_code("729f957048af4a3f") |
| get_wikitext2 | Not yet run | efficient-ml/qwen3-quantization/BiLLM/datautils.py code served (permissive licence) · get_code("011909315dfdbc2e") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
The Qwen series has emerged as a leading family of open-source Large Language Models (LLMs), demonstrating remarkable capabilities in natural language understanding tasks. With the recent release of Qwen3, which exhibits superior performance across diverse benchmarks, there is growing interest in deploying these models efficiently in resource-constrained environments. Low-bit quantization presents a promising solution, yet its impact on Qwen3's performance remains underexplored. This study conducts a systematic evaluation of Qwen3's robustness under various quantization settings, aiming to uncover both opportunities and challenges in compressing this state-of-the-art model. We rigorously assess 5 existing classic post-training quantization techniques applied to Qwen3, spanning bit-widths from 1 to 8 bits, and evaluate their effectiveness across multiple datasets. Our findings reveal that while Qwen3 maintains competitive performance at moderate bit-widths, it experiences notable degradation in linguistic tasks under ultra-low precision, underscoring the persistent hurdles in LLM compression. These results emphasize the need for further research to mitigate performance loss in extreme quantization scenarios. We anticipate that this empirical analysis will provide actionable insights for advancing quantization methods tailored to Qwen3 and future LLMs, ultimately enhancing their practicality without compromising accuracy. Our project is released on https://github.com/Efficient-ML/Qwen3-Quantization and https://huggingface.co/collections/Efficient-ML/qwen3-quantization-68164450decb1c868788cb2b.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2505.02214")
get_code_for_paper("2505.02214")
have("2505.02214")
Connect an agent — have() is free.