7.4 billion tokens per month. 34 free LLM providers. 635 free model endpoints. All behind one /v1 endpoint, plus any custom OpenAI-compatible endpoint. Smart routing, automatic failover, encrypted keys. Personal experimentation only.
TypeScript
import { Router } from 'express';
import type { Request, Response } from 'express';
import { z } from 'zod';
import { listSearchEngines, updateSearchEngine } from '../services/search.js';
// Dashboard: the Models → Search page (#1174). Engines come from the catalog's
// searchEngines registry (plus custom relays); this lists them with key and
// usage counts and lets the operator switch one off or change the order.
export const searchRouter = Router();
searchRouter.get('/engines', (_req: Request, res: Response) => {
res.json(listSearchEngines());
});
const patchSchema = z.object({
enabled: z.boolean().optional(),
priority: z.number().int().min(0).optional(),
}).refine(p => p.enabled !== undefined || p.priority !== undefined, {
message: 'Provide enabled and/or priority',
});
searchRouter.patch('/engines/:id', (req: Request, res: Response) => {
const id = Number(req.params.id);
if (!Number.isInteger(id) || id <= 0) {
res.status(400).json({ error: { message: 'Invalid engine id' } });
return;
}
const parsed = patchSchema.safeParse(req.body);
if (!parsed.success) {
res.status(400).json({ error: { message: parsed.error.errors.map(e => e.message).join(', ') } });
return;
}
if (!updateSearchEngine(id, parsed.data)) {
res.status(404).json({ error: { message: 'Search engine not found' } });
return;
}
const engine = listSearchEngines().find(e => e.id === id);
res.json(engine);
});