Zero vendor lock-in. Zero migration pain.
One SDK connects your products to LLM, Cache, Vector Search, Analytics, and more. Change providers in a config file — zero code changes.
{
"adapters": {
"llm": "openai",
"cache": "redis",
"vectorStore": "qdrant",
"analytics": "sqlite"
}
}Providers change. Your code doesn't.
Supported Adapters
LLM
Language model access with tier-based routing and automatic failover.
OpenAI, Anthropic, Ollama, Azure OpenAI
Cache
Key-value caching with TTL, sorted sets, and atomic operations.
Redis, Memcached, In-Memory
Vector Store
Semantic search over embeddings with upsert, delete, and filtering.
Qdrant, Pinecone, ChromaDB
Analytics
Event tracking, time-series aggregation, and usage dashboards.
SQLite, DuckDB, ClickHouse
Storage
File read/write/list with a unified interface across backends.
S3, Local FS, MinIO
Embeddings
Text-to-vector conversion for search and similarity.
OpenAI, Cohere, Local
import { KBPlatform } from '@kb-labs/platform-client';
const platform = new KBPlatform({
endpoint: 'http://gateway:4000',
apiKey: process.env.KB_API_KEY,
});
// LLM — provider resolved by platform config
const answer = await platform.llm.complete('Explain this code');
// Cache — Redis, Memcached, or in-memory, same API
await platform.cache.set('session:123', userData, 3600);
// Telemetry — batched, auto-flushed
platform.telemetry.event('user.signup', { plan: 'pro' });Stop integrating the same infrastructure in every product
One platform. One SDK. Every capability. Switch providers centrally, not per-product.