feat(memory): native Voyage AI support (#7078)

* feat(memory): add native Voyage AI embedding support with batching

Cherry-picked from PR #2519, resolved conflict in memory-search.ts
(hasRemote -> hasRemoteConfig rename + added voyage provider)

* fix(memory): optimize voyage batch memory usage with streaming and deduplicate code

Cherry-picked from PR #2519. Fixed lint error: changed this.runWithConcurrency
to use imported runWithConcurrency function after extraction to internal.ts
This commit is contained in:
Jake
2026-02-07 10:09:32 +13:00
committed by GitHub
parent e3d3893d5d
commit 6965a2cc9d
11 changed files with 879 additions and 58 deletions

View File

@@ -0,0 +1,86 @@
import { requireApiKey, resolveApiKeyForProvider } from "../agents/model-auth.js";
import type { EmbeddingProvider, EmbeddingProviderOptions } from "./embeddings.js";
export type VoyageEmbeddingClient = {
baseUrl: string;
headers: Record<string, string>;
model: string;
};
export const DEFAULT_VOYAGE_EMBEDDING_MODEL = "voyage-4-large";
const DEFAULT_VOYAGE_BASE_URL = "https://api.voyageai.com/v1";
export function normalizeVoyageModel(model: string): string {
const trimmed = model.trim();
if (!trimmed) return DEFAULT_VOYAGE_EMBEDDING_MODEL;
if (trimmed.startsWith("voyage/")) return trimmed.slice("voyage/".length);
return trimmed;
}
export async function createVoyageEmbeddingProvider(
options: EmbeddingProviderOptions,
): Promise<{ provider: EmbeddingProvider; client: VoyageEmbeddingClient }> {
const client = await resolveVoyageEmbeddingClient(options);
const url = `${client.baseUrl.replace(/\/$/, "")}/embeddings`;
const embed = async (input: string[]): Promise<number[][]> => {
if (input.length === 0) return [];
const res = await fetch(url, {
method: "POST",
headers: client.headers,
body: JSON.stringify({ model: client.model, input }),
});
if (!res.ok) {
const text = await res.text();
throw new Error(`voyage embeddings failed: ${res.status} ${text}`);
}
const payload = (await res.json()) as {
data?: Array<{ embedding?: number[] }>;
};
const data = payload.data ?? [];
return data.map((entry) => entry.embedding ?? []);
};
return {
provider: {
id: "voyage",
model: client.model,
embedQuery: async (text) => {
const [vec] = await embed([text]);
return vec ?? [];
},
embedBatch: embed,
},
client,
};
}
export async function resolveVoyageEmbeddingClient(
options: EmbeddingProviderOptions,
): Promise<VoyageEmbeddingClient> {
const remote = options.remote;
const remoteApiKey = remote?.apiKey?.trim();
const remoteBaseUrl = remote?.baseUrl?.trim();
const apiKey = remoteApiKey
? remoteApiKey
: requireApiKey(
await resolveApiKeyForProvider({
provider: "voyage",
cfg: options.config,
agentDir: options.agentDir,
}),
"voyage",
);
const providerConfig = options.config.models?.providers?.voyage;
const baseUrl = remoteBaseUrl || providerConfig?.baseUrl?.trim() || DEFAULT_VOYAGE_BASE_URL;
const headerOverrides = Object.assign({}, providerConfig?.headers, remote?.headers);
const headers: Record<string, string> = {
"Content-Type": "application/json",
Authorization: `Bearer ${apiKey}`,
...headerOverrides,
};
const model = normalizeVoyageModel(options.model);
return { baseUrl, headers, model };
}