Back to AIBriefs
LaunchAI Models
Featured

ReAligned-Qwen3.5 model series released to reduce ideological bias

ReAligned-Qwen3.5 model series released to reduce ideological biasView on Reddit

Finetuned from Qwen3.5, the models aim to reduce Chinese ideological bias and censorship. Released under Apache 2.0 license, trained with SFT+GRPO pipeline.

·
14 days ago
ReAligned-Qwen3.5 model series released to reduce ideological bias — AIBriefs