跳到正文
Vercel Blog(vercel.com)· Gregor Martynus·· 17 天前品質評分53

Ampersand 開放透過 Vercel AI Gateway 使用

Ampersand from Humans& now available on AI Gateway

重點導讀

Vercel AI Gateway 開放使用 Humans& 的多人對話模型 Ampersand,並提供至 2026 年 11 月 13 日的免費使用期間。

  • 訓練方式:Ampersand 以 Z.ai 的開放權重模型 GLM 5.3 為基礎,透過模擬最長數月的群體對話進行強化學習訓練。
  • 對話方式:模型會讀取每則訊息,包括未直接對它發出的訊息,並判斷要回覆或保持安靜;設計目標包括理解個人與共同目標、協助處理分歧,以及在適當時提供脈絡或提出澄清問題。
  • 接入方式:應用程式可使用呼叫端定義的 send_message 與 pass 工具,分別送出發言或等待下一則人類訊息。可透過 AI SDK、OpenAI 相容的 Chat Completions API 或程式開發 Agent 呼叫;模型名稱為 humansand/ampersand-v0.1。
  • 程式開發 Agent:在 Claude Code、Codex、Cursor 等工具中使用時,需安裝最新版 Vercel CLI 並執行設定,再選取該模型。

AI 伴隨問答

自備 Key 深度分析

針對這篇報導有疑問?可直接點選快捷問題或向 AI 追問背景脈絡與技術細節:

正文

Ampersand from Humans& is now available on AI Gateway for free through November 13, 2026.

Ampersand is a model for group conversations. Humans& built Ampersand on Z.ai's open-weight GLM 5.3 and trained it with reinforcement learning in simulated group conversations that run for up to months of simulated time. Most chat models are built for one person asking a question and the model answering it. Ampersand is trained to take part in a conversation between several people. It reads every message, including the ones not addressed to it, and after each one it decides whether to reply or stay quiet.

It is designed to understand individual and shared goals, help people work through disagreements, and recognize when a contribution would be useful. That can mean asking a clarifying question, sharing relevant context, or waiting while people respond to one another.

For example:

Sam: Heads up, I'm out Friday, driving to my sister's. Alex: Have a good trip! Maya: Let's do the launch rehearsal Friday at 10 with all three of us. Ampersand: Quick flag, Maya: Sam said he's out Friday driving to his sister's, so a Friday 10am rehearsal with all three probably won't work. Might be worth picking a different time so everyone can make it.

Your application exposes these actions as caller-defined function tools. send_message delivers a contribution, while pass leaves the conversation quiet until the next human message. Your application executes both tools and records their results.

Call it through the AI SDK, the OpenAI-compatible Chat Completions API, or a coding agent. Use humansand/ampersand-v0.1 as the model name:

To use it in Claude Code, Codex, Cursor, and more, install the latest Vercel CLI and run setup:

Then select humansand/ampersand-v0.1 in the agent. See the coding agents guide for details.

Try Ampersand in the model playground.

AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting, budgets for API keys, routing rules, and more.

AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests.

You can view all language models available on AI Gateway.

Read more

來源: Vercel Blog (vercel.com)
開啟原文

讀者討論

理性交流 · 即時討論