# AI/ML API Node.js SDK

npm License Docs

The official TypeScript/JavaScript SDK for [AI/ML API](https://aimlapi.com) - access 400+ AI models (GPT-4, Claude, Gemini, DeepSeek, etc.) with an OpenAI-compatible interface. ## Installation ```bash npm install @ai-ml.api/aimlapi-sdk-node ``` ## Quickstart ### Environment Setup Create a `.env` file with your API key: ```bash AIML_API_KEY=your-api-key-here ``` ### Import and Configure ```typescript import { AIMLAPI } from "@ai-ml.api/aimlapi-sdk-node"; import dotenv from "dotenv"; dotenv.config(); const client = new AIMLAPI(); ``` ### Run Code Snippets Use `npx tsx` to run TypeScript/JavaScript files directly: ```bash npx tsx your-script.ts ``` Or with environment file: ```bash AIML_API_KEY=your-key npx tsx your-script.ts ``` ## Available Resources - **Chat & Messages**: OpenAI-compatible chat, Anthropic messages - **Images & Vision**: Image generation, OCR, vision analysis - **Video & Audio**: Video generation, TTS, speech-to-text - **Music & Search**: Music generation, Bagoodex search - **Utilities**: Batches, embeddings, account management ## Examples ### Chat Completions ```typescript import { AIMLAPI } from "@ai-ml.api/aimlapi-sdk-node"; import dotenv from "dotenv"; dotenv.config(); const client = new AIMLAPI(); const completion = await client.chat.completions.create({ model: "gpt-4o", messages: [ { role: "system", content: "You are a helpful assistant." }, { role: "user", content: "Explain quantum computing." }, ], }); console.log(completion.choices[0]?.message?.content); ``` ### Streaming ```typescript const stream = await client.chat.completions.create({ model: "gpt-4o", messages: [{ role: "user", content: "Write a story." }], stream: true, }); for await (const chunk of stream) { process.stdout.write(chunk.choices[0]?.delta?.content ?? ""); } ``` ### Video Generation with Polling Long-running video generation with automatic status polling: ```typescript // Generate video and wait for completion const result = await client.video.createWithPolling( { model: "openai/sora-2-t2v", prompt: "A menacing evil dragon appears above a mountain", resolution: "720p", aspect_ratio: "16:9", duration: 4, }, 20, // maxAttempts 5000 // pollIntervalMs ); console.log("Video status:", result.status); console.log("Video URL:", result.video?.url); ``` Alternative: Generate first, then poll separately: ```typescript // Start generation const generation = await client.video.generate({ model: "openai/sora-2-t2v", prompt: "A robot dancing", resolution: "720p", aspect_ratio: "16:9", }); // Poll for status const status = await client.video.getStatus(generation.id); console.log("Status:", status.status); ``` ### Music Generation with Polling Music generation supports multiple models (ElevenLabs, Minimax, Google Lyria): ```typescript // Generate music and wait for completion const music = await client.music.createWithPolling( { model: "elevenlabs/eleven_music", prompt: "Lo-fi hip-hop ambient music", music_length_ms: 20000, }, 30, // maxAttempts 120 // timeoutSeconds ); console.log("Music status:", music.status); console.log("Audio URL:", music.audio_file?.url); ``` Using Minimax Music-2.0 with lyrics: ```typescript const music = await client.music.generateMusic({ model: "minimax/music-2.0", prompt: "Electronic dance music", lyrics: "[Verse]\nElectronic beats\n[Chorus]\nDance all night", audio_setting: { format: "mp3", sample_rate: 44100, }, }); console.log("Music ID:", music.id); ``` ### Speech-to-Text (STT) Transcribe audio with automatic polling for completion: ```typescript // Transcribe audio with polling const result = await client.speechToText.createWithPolling( { model: "aai/slam-1", audio: { url: "https://example.com/audio.mp3" }, }, 10, // pollInterval seconds 300 // pollTimeout seconds ); console.log("Transcription:", result.text); ``` Or submit first, then poll separately: ```typescript // Start transcription const stt = await client.speechToText.create({ model: "aai/slam-1", audio: { url: "https://example.com/audio.mp3" }, }); // Poll for status const status = await client.speechToText.getStatus(stt.generation_id); console.log("Status:", status.status); ``` ### Text-to-Speech ```typescript const audio = await client.textToSpeech.create({ model: "alibaba/qwen3-tts-flash", text: "Hello, this is a test.", voice: "Cherry", }); console.log("Audio URL:", audio.audio.url); ``` ### Image Generation ```typescript const response = await client.images.generate({ model: "dall-e-3", prompt: "A beautiful sunset over mountains", size: "1024x1024", }); console.log("Image URL:", response.data[0].url); ``` ### Account Management Get balance and available models: ```typescript // Get account balance const balance = await client.account.getBalance(); console.log("Balance:", balance.balance); console.log("Low balance:", balance.lowBalance); // Get all available models const models = await client.account.getModels(); console.log("Total models:", models.models?.length); // Filter models by type const chatModels = await client.account.getModelsByType("chat-completion"); console.log("Chat models:", chatModels.models?.length); // Filter by provider const openaiModels = await client.account.getModelsByDeveloper("openai"); console.log("OpenAI models:", openaiModels.models?.length); // Find model by ID const model = await client.account.getModelById("gpt-4o"); console.log("Model:", model); ``` ### Messages API (Anthropic-compatible) ```typescript const message = await client.messages.create({ model: "claude-sonnet-4.5", messages: [{ role: "user", content: "Hello, Claude!" }], max_tokens: 100, }); console.log(message.content[0]?.text); ``` ### OCR (Vision) ```typescript // Google Document AI OCR const ocr = await client.ocr.google({ document: "https://example.com/document.png", mimeType: "image/png", }); console.log("Pages:", ocr.pages?.length); // Mistral OCR const mistralOcr = await client.ocr.mistral({ document: { type: "image_url", image_url: "https://example.com/image.jpg", }, }); console.log("Text:", mistralOcr.pages?.[0]?.text); ``` ### Batches Process multiple requests asynchronously: ```typescript const batch = await client.batch.create({ requests: [ { custom_id: "request-1", params: { model: "claude-3-5-haiku-20241022", messages: [{ role: "user", content: "Hello" }], max_tokens: 100, }, }, ], }); console.log("Batch ID:", batch.id); console.log("Status:", batch.processing_status); // Wait for completion const result = await client.batch.waitForCompletion(batch.id, { maxAttempts: 30, pollInterval: 10, }); // Parse results const parsed = client.batch.parseResults(result as string); ``` ### Bagoodex (Search & Utilities) ```typescript // Search knowledge base const knowledge = await client.bagoodex.searchKnowledge("Who is Nikola Tesla"); console.log("Title:", knowledge.title); // Get weather const weather = await client.bagoodex.getWeather("San Francisco, CA"); console.log("Temperature:", weather.temperature); // Search images const images = await client.bagoodex.searchImages("cute cats"); console.log("Results:", images.length); // Search videos const videos = await client.bagoodex.searchVideos("tutorial"); console.log("Results:", videos.length); ``` ## Available Models | Provider | Models | |----------|--------| | OpenAI | gpt-4o, gpt-4o-mini, o1, o3-mini, gpt-5 | | Anthropic | claude-sonnet-4.5, claude-opus-4.5 | | Google | gemini-2.5-pro, gemini-2.0-flash | | DeepSeek | deepseek-chat, deepseek-reasoner | | Meta | llama-4-maverick, llama-4-scout | | Video | openai/sora-2-t2v, alibaba/wan2.1-t2v-plus, luma/ray-2 | | Music | elevenlabs/eleven_music, minimax/music-2.0, google/lyria2 | | And more... | 400+ models total | ## Environment Variables | Variable | Description | |----------|-------------| | `AIML_API_KEY` | Your API key | | `AIML_API_BASE` | Custom base URL (optional) | | `AIMLAPI_LOG` | Log level: debug, info, warn, error | ## Documentation - [API Reference](https://docs.aimlapi.com/api-reference) - [Examples](https://github.com/aimlapi/aimlapi-sdk-node/tree/main/examples) - [Python SDK](https://github.com/aimlapi/aimlapi-sdk-python) ## License Licensed under the Apache License 2.0.