> Discover all available pages from the documentation index: https://mastra.zisheng.pro/zh-TW/llms.txt # ![Venice AI 標誌](https://models.dev/logos/venice.svg)Venice AI 透過 Mastra 的模型路由器存取 25 個 Venice AI 模型。系統會自動使用 `VENICE_API_KEY` 環境變數完成身分驗證。 如需進一步了解,請參閱[Venice AI 文件](https://docs.venice.ai)。 ```bash VENICE_API_KEY=your-api-key ``` ```typescript import { Agent } from '@mastra/core/agent' const agent = new Agent({ id: 'my-agent', name: 'My Agent', instructions: 'You are a helpful assistant', model: 'venice/claude-opus-45', }) // Generate a response const response = await agent.generate('Hello!') // Stream a response const stream = await agent.stream('Tell me a story') for await (const chunk of stream) { console.log(chunk) } ``` > **資訊:** Mastra 使用與 OpenAI 相容的 `/chat/completions` endpoint。部分 Provider 特有功能可能無法使用,詳情請參閱[Venice AI 文件](https://docs.venice.ai)。 ## 模型 | Model | Context | Tools | Reasoning | Image | Audio | Video | Input $/1M | Output $/1M | | --------------------------------------- | ------- | ----- | --------- | ----- | ----- | ----- | ---------- | ----------- | | `venice/claude-opus-45` | 203K | | | | | | $6 | $30 | | `venice/claude-sonnet-45` | 203K | | | | | | $4 | $19 | | `venice/deepseek-v3.2` | 164K | | | | | | $0.40 | $1 | | `venice/gemini-3-flash-preview` | 262K | | | | | | $0.70 | $4 | | `venice/gemini-3-pro-preview` | 203K | | | | | | $3 | $15 | | `venice/google-gemma-3-27b-it` | 203K | | | | | | $0.12 | $0.20 | | `venice/grok-41-fast` | 262K | | | | | | $0.50 | $1 | | `venice/grok-code-fast-1` | 262K | | | | | | $0.25 | $2 | | `venice/hermes-3-llama-3.1-405b` | 131K | | | | | | $1 | $3 | | `venice/kimi-k2-thinking` | 262K | | | | | | $0.75 | $3 | | `venice/llama-3.2-3b` | 131K | | | | | | $0.15 | $0.60 | | `venice/llama-3.3-70b` | 131K | | | | | | $0.70 | $3 | | `venice/minimax-m21` | 203K | | | | | | $0.40 | $2 | | `venice/mistral-31-24b` | 131K | | | | | | $0.50 | $2 | | `venice/openai-gpt-52` | 262K | | | | | | $2 | $18 | | `venice/openai-gpt-52-codex` | 262K | | | | | | $2 | $18 | | `venice/openai-gpt-oss-120b` | 131K | | | | | | $0.07 | $0.30 | | `venice/qwen3-235b-a22b-instruct-2507` | 131K | | | | | | $0.15 | $0.75 | | `venice/qwen3-235b-a22b-thinking-2507` | 131K | | | | | | $0.45 | $4 | | `venice/qwen3-4b` | 33K | | | | | | $0.05 | $0.15 | | `venice/qwen3-coder-480b-a35b-instruct` | 262K | | | | | | $0.75 | $3 | | `venice/qwen3-next-80b` | 262K | | | | | | $0.35 | $2 | | `venice/qwen3-vl-235b-a22b` | 262K | | | | | | $0.25 | $2 | | `venice/venice-uncensored` | 33K | | | | | | $0.20 | $0.90 | | `venice/zai-org-glm-4.7` | 203K | | | | | | $0.55 | $3 | ## 進階設定 ### 自訂標頭 ```typescript const agent = new Agent({ id: 'custom-agent', name: 'custom-agent', model: { url: 'https://api.venice.ai/api/v1', id: 'venice/claude-opus-45', apiKey: process.env.VENICE_API_KEY, headers: { 'X-Custom-Header': 'value', }, }, }) ``` ### 動態選擇模型 ```typescript const agent = new Agent({ id: 'dynamic-agent', name: 'Dynamic Agent', model: ({ requestContext }) => { const useAdvanced = requestContext.task === 'complex' return useAdvanced ? 'venice/zai-org-glm-4.7' : 'venice/claude-opus-45' }, }) ```