Rczlin/dsh-better-reasoning ↗★ 0
@dsh-better-reasoning/plugin
Automatic thinking-level fallback for third-party (pi-ai) models + a Codex-style vertical model/effort picker for DeepSeek Harness. 适合使用第三方模型并需要手动调节或自动降级思考强度的用户。
Install
$
npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Rczlin/dsh-better-reasoningREADME
Read the full README ↗配置
cordis.patch.yml 里该插件行可加:
config:
catalogUrl: https://models.dev/api.json # 可换源
autoRefresh: true # 激活时是否后台拉取