mirror of
https://github.com/diegosouzapw/OmniRoute.git
synced 2026-09-21 22:32:22 +03:00
65e81158ab
Landed with the design call resolved per the owner's pick — **option 1**: the synced store is now endpoint-agnostic (persistDiscoveredModels and managedModelImport no longer drop non-chat models at write time), and chat selectability moved to read time (auto-pool expansion in autoStrategy applies filterChatSelectableModels; the models-route projection already had its chatOnly filter). Your discovery test now passes end-to-end (3/3): /api/show capabilities persist per connection and image/embedding requests route through the advertising host. Reconciliation notes: conflicted areas merged onto the current tip (adobe discovery import, requestedModel preflight signature, resolvedProvider fast-path coexists with the synced-route override — explicit resolution wins); carried base-red drains (#10055 memoization, #11071 test variants) dropped as already-landed; the managed-model-import exclusion test was propagated to the new contract (image/video models persist; the read filter still hides them from chat pickers — pinned by a new assertion). Full battery: 205/206 focused (the one red is a confirmed periodic-timer timing flake on the loaded devbox — 20/20 isolated), autoCombo vitest 30/30, combo suites 46/46, gates + typecheck clean. Thank you @yourspraveen — the capability probe + routing design was right; it just needed the store contract opened up. Fixes #11087.
82 lines
2.8 KiB
TypeScript
82 lines
2.8 KiB
TypeScript
import assert from "node:assert/strict";
|
|
import { describe, it, before, after } from "node:test";
|
|
import { enrichCatalogModelEntry } from "../../src/lib/modelMetadataRegistry.ts";
|
|
import {
|
|
saveModelsDevPricing,
|
|
clearModelsDevPricing,
|
|
type PricingByProvider as ModelsDevPricingByProvider,
|
|
} from "../../src/lib/modelsDevSync.ts";
|
|
import {
|
|
saveSyncedPricing,
|
|
clearSyncedPricing,
|
|
type PricingByProvider as SyncedPricingByProvider,
|
|
} from "../../src/lib/pricingSync.ts";
|
|
|
|
type CatalogPricing = {
|
|
input?: number;
|
|
output?: number;
|
|
cached?: number;
|
|
cache_creation?: number;
|
|
};
|
|
|
|
// #9364: resolveCatalogPricing() only consults models_dev_pricing and hardcoded
|
|
// defaults, skipping the LiteLLM `pricing_synced` namespace entirely. A model
|
|
// whose pricing exists ONLY in pricing_synced (the documented Layer 3) gets
|
|
// `pricing: null` in the /v1/models catalog. This test seeds pricing_synced
|
|
// with pricing for a model absent from both models_dev_pricing and hardcoded
|
|
// defaults, then asserts enrichCatalogModelEntry() surfaces it.
|
|
|
|
describe("catalog pricing LiteLLM gap (#9364)", () => {
|
|
before(() => {
|
|
// Seed ONLY the LiteLLM namespace with a model that is absent from both
|
|
// models.dev and hardcoded defaults (babbage-002 is not in default-pricing
|
|
// and not registered in the provider registry, so neither layer can match).
|
|
const synced: SyncedPricingByProvider = {
|
|
openai: {
|
|
"babbage-002": { input: 0.4, output: 0.4 },
|
|
},
|
|
};
|
|
saveSyncedPricing(synced);
|
|
|
|
// Ensure models_dev_pricing has a different model so we prove the LiteLLM
|
|
// layer is being consulted, not accidentally overlapping with models.dev.
|
|
const modelsDev: ModelsDevPricingByProvider = {
|
|
openai: {
|
|
"gpt-4o": { input: 2.5, output: 10 },
|
|
},
|
|
};
|
|
saveModelsDevPricing(modelsDev);
|
|
});
|
|
|
|
after(() => {
|
|
try {
|
|
clearSyncedPricing();
|
|
clearModelsDevPricing();
|
|
} catch {
|
|
// ignore
|
|
}
|
|
});
|
|
|
|
it("attaches LiteLLM-synced pricing onto catalog entries absent from models.dev and defaults", () => {
|
|
const entry = enrichCatalogModelEntry({
|
|
id: "openai/babbage-002",
|
|
owned_by: "openai",
|
|
root: "babbage-002",
|
|
});
|
|
assert.ok(entry.pricing, "pricing should resolve from pricing_synced (LiteLLM) layer");
|
|
assert.equal((entry.pricing as CatalogPricing).input, 0.4);
|
|
assert.equal((entry.pricing as CatalogPricing).output, 0.4);
|
|
});
|
|
|
|
it("still resolves models.dev pricing when present (precedence preserved)", () => {
|
|
const entry = enrichCatalogModelEntry({
|
|
id: "openai/gpt-4o",
|
|
owned_by: "openai",
|
|
root: "gpt-4o",
|
|
});
|
|
assert.ok(entry.pricing);
|
|
assert.equal((entry.pricing as CatalogPricing).input, 2.5);
|
|
assert.equal((entry.pricing as CatalogPricing).output, 10);
|
|
});
|
|
});
|