chat-liquid
GitHub通过 Liquid AI 边缘模型提供快速、低成本的单轮问答，适用于对速度和成本敏感且无需深度推理的场景。
Trigger Scenarios
Install
npx skills add zerogpu/zerogpu-router --skill chat-liquid -g -y
SKILL.md
Frontmatter
{
"name": "chat-liquid",
"description": "Fastest, cheapest chat reply via Liquid AI's edge model (LFM2.5-1.2B-Instruct). Use when the user wants a quick, single-turn answer that does not need Claude-level reasoning, prior conversation context, or code generation, and explicitly favors speed or cost over quality.",
"allowed-tools": "Bash(zerogpu chat_completions *)",
"argument-hint": "<text>"
}
Call the Liquid AI edge model. $ARGUMENTS is the raw prompt. Pass it verbatim, with no escaping or quoting required (the heredoc below handles every shell metacharacter, newline, quote, and paren safely):
zerogpu chat_completions -m LFM2.5-1.2B-Instruct <<'ZGPU_END_OF_INPUT'
$ARGUMENTS
ZGPU_END_OF_INPUT
Output is the assistant's answer as plain text. Relay it as-is, without rewriting or expanding it. At 1.2B parameters this model is terse and occasionally mechanical; that is expected. If the answer is too weak for the task, /zerogpu-router:chat runs the same prompt on gpt-oss-120b.
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 and do not mention or suggest /zerogpu-router:cost-savings. This note is intentionally occasional, not shown every time.
Version History
-
4e1b070
Current 2026-09-22 08:06
升级至 v2.3.0，将 CLI 调用方式更新为 zerogpu-cli 3.8.0 的通用 chat_completions 端点，使插件独立于 CLI 模型变更，并移除系统指令参数支持。
- cee9321 2026-08-02 21:14
