360 Secure SkillHub
360 安全技能中心

chat-thinking

通过 Bash 执行 zerogpu chat_thinking 命令,把用户输入经 heredoc 原样传给 ZeroGPU 的 LFM2.5-1.2B-Thinking 模型,适用于需要展示逐步推理的简短逻辑、数学或应用题;输出包含模型的推理轨迹,若输出含节省提示行则原样追加到回复末尾。

zerogpu
zerogpu
热度 13.63
最近更新 2026-07-14 04:17:56
分类:知识依赖:代码执行 / 第三方 API / 内部调用大模型应用场景:概念解释 / 教学辅导

安装技能

复制下面的话术发给你的 Agent,它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill: npx secure-skills-cli add chat-thinking
Metadata
name
chat-thinking
description
Chat with ZeroGPU's Thinking variant (LFM2.5-1.2B-Thinking), which returns visible reasoning alongside the answer. Use when the user wants the model's reasoning shown, or asks a short logic/math/word-problem question that benefits from step-by-step output.
argument-hint
<text>
allowed-tools
Bash(zerogpu chat_thinking*)

Call the ZeroGPU thinking model. $ARGUMENTS is the raw user prompt — pass it verbatim, no escaping or quoting required (the heredoc below handles every shell metacharacter, newline, quote, and paren safely):

ZGPU_TEXT=$(cat <<'ZGPU_END_OF_INPUT'
$ARGUMENTS
ZGPU_END_OF_INPUT
)
zerogpu chat_thinking "$ZGPU_TEXT"

The response includes the model's reasoning trace.

Savings note: only if the command output literally contains a line starting with 💰 ZeroGPU savings, append that exact line, unchanged, as the last line of your reply. If no such line is present, say nothing about savings.