%20--%3e%3cdefs%3e%3cstyle%3e%20.st0%20{%20fill:%20%23061b40;%20}%20.st1%20{%20fill:%20%23306af1;%20}%20.st2%20{%20fill:%20%235ce5cf;%20}%20%3c/style%3e%3c/defs%3e%3cg%3e%3cpath%20class='st0'%20d='M55,10.5h9v3h-9v9h12V7.5h-12v3ZM64,19.5h-6v-3h6v3Z'/%3e%3cpolygon%20class='st0'%20points='69%2016.5%2078%2016.5%2078%2019.5%2069%2019.5%2069%2022.5%2081%2022.5%2081%2013.5%2072%2013.5%2072%2010.5%2081%2010.5%2081%207.5%2069%207.5%2069%2016.5'/%3e%3cpolygon%20class='st0'%20points='95%2010.5%2095%207.5%2083%207.5%2083%2022.5%2095%2022.5%2095%2019.5%2086%2019.5%2086%2016.5%2095%2016.5%2095%2013.5%2086%2013.5%2086%2010.5%2095%2010.5'/%3e%3cpath%20class='st0'%20d='M40,1.5v21h11.6l1.4-1.4v-7.6h0c0,0-1.4-1.5-1.4-1.5l1.4-1.4V2.9l-1.4-1.4h-11.6ZM50,19.5h-7v-6h7v6ZM50,10.5h-7v-6h7v6Z'/%3e%3c/g%3e%3cpath%20class='st1'%20d='M23.1,24L14.7,4.8l-4.9,11.2h3.8l-1.8,4H3.3L12.1,0H2C.9,0,0,.9,0,2v20c0,1.1.9,2,2,2h21.1Z'/%3e%3cpath%20class='st2'%20d='M34,0h-16.8l10.6,24h6.2c1.1,0,2-.9,2-2V2C36,.9,35.1,0,34,0ZM32.5,20h-4V4h4v16Z'/%3e%3c/svg%3e)
Smoothie Qwen3 4B
Smoothie Qwen is a lightweight adjustment tool that can smooth the token probabilities in Qwen and similar models and enhance the multilingual balanced generation ability.
Downloads 2,189
Release Time : 4/30/2025
Model Overview
Smoothie Qwen optimizes the generation effect of Qwen series models in a multilingual environment by adjusting the token probability distribution, and is specially optimized for specific Unicode character ranges such as Chinese.
Model Features
Token probability smoothing
Adjust the token probability distribution through the minimum scaling factor and smoothness parameter to improve the generation quality
Multilingual optimization
Specifically optimized for 11 Chinese-related Unicode ranges to enhance Chinese generation ability
Lightweight adjustment
Implemented as an additional tool layer without changing the base model structure to maintain efficient operation
N-gram weight control
Supports configuring the weight ratios of different N-grams (default [0.5, 0.3, 0.2]) to finely control the generation effect
Model Capabilities
Text generation optimization
Multilingual balanced generation
Chinese text enhancement
Probability distribution adjustment
Use Cases
Content generation
Multilingual content creation
Generate balanced multilingual text content to avoid dominance of specific languages
Improve the generation quality of non-English languages
Chinese text optimization
Generate text content more in line with Chinese expression habits
Cover 26,153 target Chinese tokens
Model tuning
Enhancement of Qwen series models
Used as an additional optimization layer for Qwen series models
Modify the probabilities of 27,564 tokens
Featured Recommended AI Models
Š 2025AIbase