AI & ML
Intelligence
on demand.
42 AI capabilities accessible through the ee.ai face. Any model, any modality, one SDK call. No API keys to manage, no infrastructure to provision.
42
AI capabilities
12
Model providers
97
Languages supported
<200ms
Time to first token
Capabilities
LLM Inference
Run any language model — Claude, GPT, Llama, Mistral — through a unified API with automatic fallback.
Embeddings
Generate vector embeddings for text, images, and multimodal inputs. Optimized for semantic search and clustering.
RAG Pipelines
Retrieval-augmented generation with built-in chunking, indexing, and context window management.
AI Agents
Autonomous agents with tool use, memory, and multi-step reasoning. Deploy as serverless functions.
Vision
Image understanding, OCR, object detection, and visual Q&A. Any image format, any resolution.
Speech-to-Text
Real-time and batch transcription across 97 languages. Speaker diarization and punctuation included.
Text-to-Speech
Natural voice synthesis with 32 voices, SSML support, and streaming audio output.
Summarization
Extractive and abstractive summarization for documents, conversations, and structured data.
Classification
Zero-shot and fine-tuned classification for text, sentiment, intent, and custom taxonomies.
Entity Extraction
Named entity recognition, relation extraction, and structured data extraction from unstructured text.
Code Example
Streaming inference in five lines.
import { ee } from "@vertebase/sdk";
// Stream a response from any supported model
const stream = await ee.ai.generate({
model: "claude-3.5-sonnet",
prompt: "Explain quantum computing to a five-year-old",
max_tokens: 2048,
stream: true,
});
for await (const chunk of stream) {
process.stdout.write(chunk.text);
}
// => "Imagine you have a magic coin that can be heads AND tails..."Ready to add intelligence?
Start calling AI capabilities with the SDK — no API keys, no setup.
Get the SDK