chat-deepseek-v4-1-flash
GitHub调用 DeepSeek v4.1-flash 模型进行长上下文、高难度推理及函数调用任务。适用于大型代码库和复杂多步 Agent 工作流，提供平文本输出并附带成本节省提示。
触发场景
安装
npx skills add zerogpu/zerogpu-router --skill chat-deepseek-v4-1-flash -g -y
SKILL.md
Frontmatter
{
"name": "chat-deepseek-v4-1-flash",
"description": "Chat with deepseek-v4.1-flash, a sparse MoE model on DeepSeek's Causal Encoder-Decoder architecture (8B active on input, 16B on output) with a 1M-token context window. Use for large codebases, long documents, extended conversations, and multi-step agent tasks that want higher-effort reasoning or function calling. Costs twice chat on input and output alike, over eight times its context.",
"allowed-tools": "Bash(zerogpu chat_completions *)",
"argument-hint": "<text>"
}
Call deepseek-v4.1-flash. $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 deepseek-v4.1-flash <<'ZGPU_END_OF_INPUT'
$ARGUMENTS
ZGPU_END_OF_INPUT
At $0.30 / $1.20 per 1M input/output tokens this is the middle of the three 1M-context models on the platform — three times /zerogpu-router:chat-glm-5-3-flash on input and over three times on output, and roughly three quarters of /zerogpu-router:chat-gpt-4-1-mini on both. Prefer it when the task wants V4.1's higher-effort reasoning or function calling; for ordinary coding and agentic work at the same context size, chat-glm-5-3-flash is cheaper.
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
更新定价为输入$0.30/输出$1.20 per 1M tokens，并重新计算了与其他模型的性价比比例。
-
59e0f53
2026-09-27 19:55
价格调整：输入成本降至约八分之一，输出成本降至约六分之一；新增与 chat-glm-5-3-flash 的性能对比建议。
- 4e1b070 2026-09-22 08:06
