Make your agents perform at their peak. Slash latency and get sub-second responses with a multi-layer caching engine designed for agentic workloads.
import OpenAI from "openai";
import { Alchymos } from '@alchymos/openai';
const openai = new OpenAI(process.env.OPENAI_API_KEY);
const alchymos = new Alchymos(process.env.ALCHYMOS_API_KEY);
alchymos.withLlm(openai);
const response = await openai.chat.completions.create(params);Platform capabilities
Caching, observability, and integrations for teams building and operating AI agents.
Integrate with a lightweight SDK and keep your existing application architecture.
Reuse repeated model and tool responses to reduce unnecessary upstream requests.
Manage inference spend by serving eligible repeated requests from cache.
Distribute cached responses across regions to serve users closer to the edge.
Review workflow activity, cache behavior, latency, and usage in one place.
Use TypeScript-first integrations with typed configuration and familiar client APIs.
Connect model and tool requests to caching, delivery, and runtime visibility through a focused integration.
Layered Edge Cache
A multi-layer cache for model and tool requests. Configure reuse behavior to reduce repeated work while preserving control over relevance.
Monitor cache effectiveness and performance in real-time with detailed hit rate analytics and bandwidth savings.
Configure exactly what gets cached and for how long. Create rules based on provider, model, user, or custom metadata.
Define precise rules with TTLs, priority levels, and custom tags for powerful cache control.
Cache based on meaning, not just exact string matches.
Workflow observability
Stop flying blind. Alchymos gives you deep visibility into your AI traffic. Track costs, debug latency spikes, and audit agent decisions in real-time.
Track token usage, latency, and error rates by user, team, or agent. Drill down into specific timeframes to identify usage patterns and anomalies.
Monitor request latency and cache hit/miss durations in real time. Drill into p50 and p90 response times, identify cache misses, and correlate upstream latency to improve agent performance.
Average duration (ms) for cache hits and misses
Debug failures by replaying exact request sequences. Identify cache misses and optimize your prompt chains with full visibility into every request step.
Usage and cost
Review token consumption and cache reuse in one focused view. Connect repeated work to its effect on billed usage without switching between separate dashboards.
Original and billed usage by day
Framework integrations
Drop-in compatibility with the most popular AI frameworks and providers. No complex migration required. Just change the base URL or use our lightweight SDK.
import OpenAI from "openai";
import { Alchymos } from '@alchymos/openai';
const openai = new OpenAI(process.env.OPENAI_API_KEY);
const alchymos = new Alchymos(process.env.ALCHYMOS_API_KEY);
alchymos.withLlm(openai);
const response = await openai.chat.completions.create(params);Connect OpenAI compatible clients, MCP tools, and the Vercel AI SDK. Python support is in development; custom integrations are available on request.
TypeScript
TypeScript
TypeScript
Python (coming soon)
Python (coming soon)
Any Language (upon request)
Get started
Add caching and observability to your AI workflows with a focused integration.