minpeter/Qwen3-0.6B-Instruct

Qwen3-0.6B (Modified)

A fork of Qwen/Qwen3-0.6B, modified for use as a training target model with PrimeIntellect-ai/verifiers.

Changes

• Extracted chat_template from tokenizer_config.json into a separate chat_template.jinja file (latest transformers format)
• Reversed thinking tag logic to enable thinking mode by default (enable_thinking=True)

Original Model

For model architecture, performance, and usage details, refer to Qwen/Qwen3-0.6B.