chat-gpt-5-6-luna
GitHub调用 gpt-5.6-luna 模型进行文本生成，适用于编程、聊天、推理、RAG 及翻译等场景。支持长上下文与结构化输出，侧重成本优化，直接返回纯文本结果。
触发场景
安装
npx skills add zerogpu/zerogpu-router --skill chat-gpt-5-6-luna -g -y
SKILL.md
Frontmatter
{
"name": "chat-gpt-5-6-luna",
"description": "Chat with gpt-5.6-luna, the cost-optimized model of OpenAI's GPT-5.6 family, with a 272K-token context window, adjustable reasoning effort, function calling, and structured outputs. Use for cost-sensitive, high-volume coding, chat, reasoning, RAG, summarization, and translation work. Costs twice chat-gpt-6-luna on input and over twice on output, over the same 272K window.",
"allowed-tools": "Bash(zerogpu chat_completions *)",
"argument-hint": "<text>"
}
Call gpt-5.6-luna. $ARGUMENTS is the raw prompt — pass it verbatim, no escaping or quoting required (the heredoc below handles every shell metacharacter, newline, quote, and paren safely):
zerogpu chat_completions -m gpt-5.6-luna <<'ZGPU_END_OF_INPUT'
$ARGUMENTS
ZGPU_END_OF_INPUT
At $0.20 / $1.20 per 1M input/output tokens this costs twice /zerogpu-router:chat-gpt-6-luna on input and over twice on output, over the same 272K-token window, and a third more than /zerogpu-router:chat on input with twice its output price. For a prompt that fits in 131K tokens, /zerogpu-router:chat is cheaper on both.
Output is the assistant's answer as plain text — the model's reasoning trace comes back in a separate field and is not printed. Relay the answer as-is — do not rewrite or expand it.
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.
版本历史
-
605eceb
当前 2026-11-02 19:12
更新价格对比基准，将 glm-5.2 比较重写为 gpt-6-luna 和 chat 模型；同步仪表盘 API 版本至 v3.3.0。
- 59e0f53 2026-09-27 19:56
