feat(provider): add OpenVecta AI inference gateway (#6833)

* fix(electron): bump electron 42→43 + build better-sqlite3 from source (ABI 148) (#6605) fix(electron): bump electron 42→43 + rebuild better-sqlite3 from source against the Electron ABI (148). Electron 43 raises NODE_MODULE_VERSION to 148; better-sqlite3@12.11.1 has no electron-v148 prebuild, so the packaged app died with 'Nenhum driver SQLite disponível'. prepare-electron-standalone now compiles better-sqlite3 from source against the electron headers into build/Release (where 'bindings' resolves it). Validated by Electron Package Smoke (green) + local (node_register_module_v148). Supersedes #6378. (--admin: the only reds are SonarQube/SonarCloud failing on a coverage-report artifact digest-mismatch — a GitHub Actions infra flake, not this diff; Sonar is green on main and the diff touches only the electron build.) * deps: bump the development group across 1 directory with 6 updates (#6588) deps: bump the development group (6 updates). Rebased onto current main; all checks green after the electron-smoke fix (#6605). * fix(proxy): force CONNECT tunnel for HTTP proxied requests (undici 8.7) + production deps bump (#6620) fix(proxy): force CONNECT tunnel for HTTP proxied requests (undici 8.7) + production deps bump. undici 8.6+ changed ProxyAgent to forward plain-HTTP via request-proxy instead of CONNECT, breaking OAuth refresh through a connection proxy (501). proxyDispatcher now passes proxyTunnel:true. Validated: Unit Tests 3/8 (the OAuth-proxy test) green, new regression test green (fails without the fix on undici 8.7), SonarQube green. Supersedes #6380. (--admin: the only red is Electron Package Smoke failing on a next-build artifact 'digest-mismatch' — a GitHub Actions infra flake corrupting the asar ('file data stream has unexpected number of bytes'); the better-sqlite3 rebuild itself succeeded (gyp ok) and the electron path is unchanged from #6605 which passed the smoke. Not this diff.) * feat(provider): add OpenVecta AI inference gateway OpenVecta (https://openvecta.com/) is an OpenAI-compatible AI inference gateway hosting LLMs (GLM, Claude, DeepSeek, GPT OSS, Llama, Kimi, Nemotron...) plus text-embedding-* models behind a single Bearer key. Wiring (7 integration points): - src/shared/constants/providers/apikey/inference-hosts.ts: catalog entry - open-sse/config/providers/registry/openvecta/index.ts: registry w/ 9 seed LLMs - open-sse/config/providers/index.ts: wire into REGISTRY - src/app/api/providers/[id]/models/discovery/providerModelsConfig.ts: live /v1/models URL - src/app/api/providers/[id]/models/discovery/providerSets.ts: NAMED_OPENAI_STYLE_PROVIDERS - public/providers/openvecta.svg: brand icon - tests/unit/openvecta-provider-registration.test.ts: regression guard (6 tests, all pass) No executor needed — buildOpenAiCompatibleRegistryEntry wires format=openai / executor=default / authType=apikey / authHeader=bearer. Live catalog discovery uses the existing NAMED_OPENAI_STYLE_PROVIDERS path (live /v1/models fetch + registry seed as offline fallback). Validation: - npm run typecheck:core clean - npm run typecheck:noimplicit:core 4 errors in unchanged files (combo.ts, cliRuntime.ts); 0 in new code - npm run lint clean - node --import tsx/esm --test tests/unit/openvecta-provider-registration.test.ts 6/6 pass - sibling tests/unit/openai-style-providers-4239-4155-3841.test.ts 18/18 pass (no regression) * chore(merge): drop unrelated main-drift from PR fork + fix count/golden drift The fork branch predated main's electron 42→43 bump (#6605) and several other package.json/lockfile churn; those files are unrelated to the OpenVecta provider addition and were reintroducing an older/stale state (version 3.8.46, electron 42, older bun/eslint-config-next) that broke the Electron Package Smoke check. Restored package.json, package-lock.json, electron/package.json, electron/package-lock.json, and scripts/build/prepare-electron-standalone.mjs to match origin/release/v3.8.47. Also updates the two provider-count assertions (166->167) and regenerates the translate-path golden snapshot to account for the new openvecta entry. Co-authored-by: hajilok <120608486+hajilok@users.noreply.github.com> --------- Co-authored-by: Diego Rodrigues de Sa e Souza <8016841+diegosouzapw@users.noreply.github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: diegosouzapw <diegosouzapw@users.noreply.github.com> Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouza.pw@gmail.com>
2026-08-07 15:52:52 +03:00 · 2026-07-12 11:58:53 +07:00
parent 84c437d19e
commit 6530c92aa6
9 changed files with 226 additions and 7 deletions
--- a/open-sse/config/providers/index.ts
+++ b/open-sse/config/providers/index.ts
@@ -105,6 +105,7 @@ import { uncloseaiProvider } from "./registry/uncloseai/index.ts";
 import { nscaleProvider } from "./registry/nscale/index.ts";
 import { chatgpt_webProvider } from "./registry/chatgpt-web/index.ts";
 import { openrouterProvider } from "./registry/openrouter/index.ts";
+import { openvectaProvider } from "./registry/openvecta/index.ts";
 import { orcarouterProvider } from "./registry/orcarouter/index.ts";
 import { copilot_webProvider } from "./registry/copilot-web/index.ts";
 import { copilot_m365_webProvider } from "./registry/copilot-m365-web/index.ts";
@@ -290,6 +291,7 @@ export const REGISTRY: Record<string, RegistryEntry> = {
  nscale: nscaleProvider,
  "chatgpt-web": chatgpt_webProvider,
  openrouter: openrouterProvider,
+  openvecta: openvectaProvider,
  orcarouter: orcarouterProvider,
  "copilot-web": copilot_webProvider,
  "copilot-m365-web": copilot_m365_webProvider,
--- a/open-sse/config/providers/registry/openvecta/index.ts
+++ b/open-sse/config/providers/registry/openvecta/index.ts
@@ -0,0 +1,30 @@
+import type { RegistryEntry } from "../../shared.ts";
+import { buildOpenAiCompatibleRegistryEntry } from "../../shared.ts";
+
+/**
+ * OpenVecta — OpenAI-compatible AI inference gateway (https://openvecta.com/).
+ *
+ * `GET /v1/models` returns the live catalog (LLMs + text-embedding-* models) and
+ * is the source of truth at runtime via NAMED_OPENAI_STYLE_PROVIDERS. The seed
+ * models below cover the most-used LLMs as the offline fallback when the live
+ * fetch fails (network/auth) — same pattern as Together AI / Cerebras / NVIDIA NIM.
+ *
+ * `contextLength` is taken from the upstream `context_length` field per model
+ * (verified live via the OpenVecta /v1/models endpoint, 2026-07-11).
+ */
+export const openvectaProvider: RegistryEntry = buildOpenAiCompatibleRegistryEntry({
+  id: "openvecta",
+  alias: "openvecta",
+  baseUrl: "https://api.openvecta.com/v1/chat/completions",
+  models: [
+    { id: "glm-4.7-flash", name: "GLM 4.7 Flash", contextLength: 131072 },
+    { id: "claude-sonnet-4.6", name: "Claude Sonnet 4.6", contextLength: 1000000 },
+    { id: "deepseek-v4-flash", name: "DeepSeek V4 Flash", contextLength: 131072 },
+    { id: "gpt-oss-120b", name: "GPT OSS 120B", contextLength: 131072 },
+    { id: "gemma-4-31b", name: "Gemma 4 31B", contextLength: 262144 },
+    { id: "kimi-k2.6", name: "Kimi K2.6", contextLength: 200000 },
+    { id: "llama-3.3-70b-instruct", name: "Llama 3.3 70B Instruct", contextLength: 131072 },
+    { id: "llama-4-maverick", name: "Llama 4 Maverick", contextLength: 1048576 },
+    { id: "nemotron-3-super-120b", name: "Nemotron 3 Super 120B", contextLength: 262144 },
+  ],
+});
--- a/public/providers/openvecta.svg
+++ b/public/providers/openvecta.svg
@@ -0,0 +1,12 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 64 64" width="64" height="64" role="img" aria-label="OpenVecta">
+  <title>OpenVecta</title>
+  <defs>
+    <linearGradient id="ov-grad" x1="0" y1="0" x2="1" y2="1">
+      <stop offset="0%" stop-color="#7C3AED" />
+      <stop offset="100%" stop-color="#A855F7" />
+    </linearGradient>
+  </defs>
+  <rect x="2" y="2" width="60" height="60" rx="14" fill="url(#ov-grad)" />
+  <path d="M14 18 L32 46 L50 18" fill="none" stroke="#FFFFFF" stroke-width="6" stroke-linecap="round" stroke-linejoin="round" />
+  <circle cx="32" cy="40" r="3.5" fill="#FFFFFF" />
+</svg>
--- a/src/app/api/providers/[id]/models/discovery/providerModelsConfig.ts
+++ b/src/app/api/providers/[id]/models/discovery/providerModelsConfig.ts
@@ -270,6 +270,18 @@ export const PROVIDER_MODELS_CONFIG: Record<string, ProviderModelsConfigEntry> =
    authPrefix: "Bearer ",
    parseResponse: (data) => data.data || data.models || [],
  },
+  // OpenVecta (https://openvecta.com/) — OpenAI-compatible `/v1/models` returning
+  // { object: "list", data: [{ id, context_length, owned_by, … }, …] }. Bearer
+  // token with the `ov_sk_…` prefix. Same discovery shape as Together AI /
+  // Cerebras / NVIDIA NIM (live-fetch path; registry seed is the offline fallback).
+  openvecta: {
+    url: "https://api.openvecta.com/v1/models",
+    method: "GET",
+    headers: { "Content-Type": "application/json" },
+    authHeader: "Authorization",
+    authPrefix: "Bearer ",
+    parseResponse: (data) => data.data || data.models || [],
+  },
  fireworks: {
    url: "https://api.fireworks.ai/inference/v1/models",
    method: "GET",
--- a/src/app/api/providers/[id]/models/discovery/providerSets.ts
+++ b/src/app/api/providers/[id]/models/discovery/providerSets.ts
@@ -64,6 +64,12 @@ export const NAMED_OPENAI_STYLE_PROVIDERS = new Set([
  // it proxies many upstream models (DeepSeek, Claude, Kimi...) behind one key, so the
  // full catalog is discovered live from https://api.qnaigc.com/v1/models.
  "qiniu",
+  // OpenVecta (https://openvecta.com/) is an OpenAI-compatible AI inference gateway
+  // hosting LLMs (GLM, Claude, DeepSeek, GPT OSS, Llama, Kimi, Nemotron…) plus
+  // text-embedding-* models behind a single Bearer key. The full catalog is
+  // discovered live from https://api.openvecta.com/v1/models; the registry seed
+  // (registry/openvecta) covers the most-used LLMs as the offline fallback.
+  "openvecta",
 ]);

 export function isNamedOpenAIStyleProvider(provider: string): boolean {
--- a/src/shared/constants/providers/apikey/inference-hosts.ts
+++ b/src/shared/constants/providers/apikey/inference-hosts.ts
@@ -17,6 +17,20 @@ export const APIKEY_PROVIDERS_INFERENCE = {
      apiKeyUrl: "https://api.together.ai/settings/api-keys",
    },
  },
+  // OpenVecta — OpenAI-compatible AI inference gateway (https://openvecta.com/).
+  // Catalog seeded from the live /v1/models list (LLMs + embeddings); free credits
+  // advertised on signup. Bearer-token auth via Authorization: Bearer ov_sk_…
+  openvecta: {
+    id: "openvecta",
+    alias: "openvecta",
+    name: "OpenVecta",
+    icon: "vector_polygon",
+    color: "#7C3AED",
+    textIcon: "OV",
+    website: "https://openvecta.com",
+    hasFree: true,
+    freeNote: "Free credits on signup for OpenAI-compatible inference across LLMs, embeddings, and reasoning models",
+  },
  fireworks: {
    id: "fireworks",
    alias: "fireworks",
--- a/tests/snapshots/provider/translate-path.json
+++ b/tests/snapshots/provider/translate-path.json
@@ -3245,6 +3245,29 @@
      "stream": "https://openrouter.ai/api/v1/chat/completions"
    }
  },
+  "openvecta": {
+    "format": "openai",
+    "headers": {
+      "apiKey": {
+        "Accept": "text/event-stream",
+        "Authorization": "Bearer <TOK>",
+        "Content-Type": "application/json"
+      },
+      "nonStream": {
+        "Authorization": "Bearer <TOK>",
+        "Content-Type": "application/json"
+      },
+      "oauth": {
+        "Accept": "text/event-stream",
+        "Authorization": "Bearer <TOK>",
+        "Content-Type": "application/json"
+      }
+    },
+    "url": {
+      "nonStream": "https://api.openvecta.com/v1/chat/completions",
+      "stream": "https://api.openvecta.com/v1/chat/completions"
+    }
+  },
  "orcarouter": {
    "format": "openai",
    "headers": {
--- a/tests/unit/openvecta-provider-registration.test.ts
+++ b/tests/unit/openvecta-provider-registration.test.ts
@@ -0,0 +1,119 @@
+/**
+ * Coverage for the OpenVecta (https://openvecta.com/) provider registration.
+ *
+ * Validates the seven wiring points established in the feat/openvecta-provider
+ * branch:
+ *   1. APIKEY_PROVIDERS.openvecta — catalog entry (id, alias, name, website, hasFree)
+ *   2. providerRegistry.openvecta — format=openai / executor=default / apikey / bearer
+ *   3. PROVIDER_ENDPOINTS — chat completions baseUrl
+ *   4. providerModelsConfig — live /v1/models discovery URL
+ *   5. NAMED_OPENAI_STYLE_PROVIDERS — openvecta classified for live-fetch
+ *   6. Seeded registry catalog — non-empty, unique ids, includes a curated LLM subset
+ *   7. Schema validation — Zod validateProviders(APIKEY_PROVIDERS) passes (the import
+ *      would throw at module-load time if any entry were malformed)
+ *
+ * Rule #18 (TDD/VPS gate) — this test exists to keep the seven wiring points
+ * in sync; if any one drifts, the relevant assertion fails immediately.
+ */
+import test from "node:test";
+import assert from "node:assert/strict";
+
+const { APIKEY_PROVIDERS } = await import(
+  "../../src/shared/constants/providers.ts"
+);
+const { REGISTRY: providerRegistry } = await import(
+  "../../open-sse/config/providerRegistry.ts"
+);
+const { NAMED_OPENAI_STYLE_PROVIDERS, isNamedOpenAIStyleProvider } = await import(
+  "../../src/app/api/providers/[id]/models/discovery/providerSets.ts"
+);
+const { PROVIDER_MODELS_CONFIG } = await import(
+  "../../src/app/api/providers/[id]/models/discovery/providerModelsConfig.ts"
+);
+
+const SPEC = {
+  id: "openvecta",
+  alias: "openvecta",
+  name: "OpenVecta",
+  website: "https://openvecta.com",
+  chatUrl: "https://api.openvecta.com/v1/chat/completions",
+  modelsUrl: "https://api.openvecta.com/v1/models",
+  expectedSeedIds: [
+    "glm-4.7-flash",
+    "claude-sonnet-4.6",
+    "deepseek-v4-flash",
+    "gpt-oss-120b",
+    "gemma-4-31b",
+    "kimi-k2.6",
+    "llama-3.3-70b-instruct",
+    "llama-4-maverick",
+    "nemotron-3-super-120b",
+  ],
+};
+
+test("APIKEY_PROVIDERS.openvecta is registered with the canonical identity", () => {
+  const entry = APIKEY_PROVIDERS[SPEC.id];
+  assert.ok(entry, `APIKEY_PROVIDERS.${SPEC.id} must be defined`);
+  assert.equal(entry.id, SPEC.id);
+  assert.equal(entry.alias, SPEC.alias);
+  assert.equal(entry.name, SPEC.name);
+  assert.equal(entry.website, SPEC.website);
+  assert.equal(typeof entry.textIcon, "string");
+  assert.equal(entry.hasFree, true, "OpenVecta advertises free signup credits");
+  assert.equal(typeof entry.freeNote, "string");
+  assert.match(entry.color, /^#[0-9A-Fa-f]{6}$/);
+});
+
+test("providerRegistry exposes the OpenAI-compatible chat completions URL", () => {
+  // Note: `PROVIDER_ENDPOINTS` in src/shared/constants/config.ts is a hand-curated
+  // *display-only* map (a subset of providers shown in the UI), so we don't
+  // require openvecta to be listed there — the source-of-truth chat URL lives
+  // on the registry entry, generated by `generateLegacyProviders()`.
+  assert.equal(providerRegistry[SPEC.id].baseUrl, SPEC.chatUrl);
+});
+
+test("PROVIDER_MODELS_CONFIG exposes the live /v1/models discovery URL", () => {
+  const cfg = PROVIDER_MODELS_CONFIG[SPEC.id];
+  assert.ok(cfg, `PROVIDER_MODELS_CONFIG.${SPEC.id} must be defined`);
+  assert.equal(cfg.url, SPEC.modelsUrl);
+  assert.equal(cfg.method, "GET");
+  assert.equal(cfg.authHeader, "Authorization");
+  assert.equal(cfg.authPrefix, "Bearer ");
+  assert.equal(typeof cfg.parseResponse, "function");
+});
+
+test("providerRegistry.openvecta uses OpenAI format with bearer apikey auth", () => {
+  const entry = providerRegistry[SPEC.id];
+  assert.ok(entry, `providerRegistry.${SPEC.id} must be defined`);
+  assert.equal(entry.id, SPEC.id);
+  assert.equal(entry.alias, SPEC.alias);
+  assert.equal(entry.format, "openai");
+  assert.equal(entry.executor, "default");
+  assert.equal(entry.authType, "apikey");
+  assert.equal(entry.authHeader, "bearer");
+  assert.equal(entry.baseUrl, SPEC.chatUrl);
+});
+
+test("openvecta is classified as a named OpenAI-style provider (live-fetch path)", () => {
+  assert.ok(
+    NAMED_OPENAI_STYLE_PROVIDERS.has(SPEC.id),
+    "openvecta must be in NAMED_OPENAI_STYLE_PROVIDERS for live /v1/models fetch",
+  );
+  assert.equal(isNamedOpenAIStyleProvider(SPEC.id), true);
+});
+
+test("openvecta ships a non-empty unique seed catalog covering curated LLMs", () => {
+  const models = providerRegistry[SPEC.id].models;
+  assert.ok(Array.isArray(models), "registry models must be an array");
+  assert.ok(models.length >= 5, "seed list must be non-empty for the offline fallback");
+  const ids = models.map((m: { id: string }) => m.id);
+  assert.equal(new Set(ids).size, ids.length, "seed model ids must be unique");
+  for (const expected of SPEC.expectedSeedIds) {
+    assert.ok(ids.includes(expected), `seed list must include ${expected}`);
+  }
+  for (const m of models) {
+    assert.equal(typeof m.id, "string");
+    assert.ok(m.id.length > 0);
+    assert.equal(typeof m.name, "string");
+  }
+});
--- a/tests/unit/providers-constants-split.test.ts
+++ b/tests/unit/providers-constants-split.test.ts
@@ -1,12 +1,13 @@
 // Characterization of the providers.ts catalog split (god-file decomposition): the host became a
 // barrel that re-exports 10 data catalogs now living under constants/providers/*, and APIKEY is
 // merged from 6 semantic family files (apikey/<family>.ts). Locks: the public surface (every catalog
-// + helpers still exported), the spread-merge integrity (166 APIKEY entries, no loss/dup), and that
+// + helpers still exported), the spread-merge integrity (167 APIKEY entries, no loss/dup), and that
 // load-time Zod validation still runs. Pure-data move → behavior must be identical.
 // Count was 171 before obsolete provider removals (PR #6675: glhf/kluster/cablyai/inclusionai etc.,
 // 171->167) plus #6126 (ClinePass dual-auth): the API-key-only APIKEY_PROVIDERS_GATEWAYS entry was
 // removed as a duplicate now that clinepass is OAuth-primary (OAUTH_PROVIDERS.clinepass) with its
-// BYOK path admitted through the DUAL_AUTH_APIKEY_PROVIDER_IDS gate instead (167->166).
+// BYOK path admitted through the DUAL_AUTH_APIKEY_PROVIDER_IDS gate instead (167->166), then the
+// OpenVecta inference-gateway addition brought it back to 167.
 import { test } from "node:test";
 import assert from "node:assert/strict";

@@ -35,12 +36,12 @@ test("barrel still exports every catalog + key helpers", () => {
  }
 });

-test("APIKEY_PROVIDERS merges the 6 family files into 166 entries (no loss / no dup)", async () => {
+test("APIKEY_PROVIDERS merges the 6 family files into 167 entries (no loss / no dup)", async () => {
  const keys = Object.keys((P as Record<string, object>).APIKEY_PROVIDERS);
-  assert.equal(keys.length, 166);
-  assert.equal(new Set(keys).size, 166, "duplicate keys after spread-merge");
+  assert.equal(keys.length, 167);
+  assert.equal(new Set(keys).size, 167, "duplicate keys after spread-merge");
  // the merged object's entry-count equals the sum of the 6 semantic family files; families are a
-  // strict partition (every provider in exactly one), so the sum must be exactly 166.
+  // strict partition (every provider in exactly one), so the sum must be exactly 167.
  const families: [string, string][] = [
    ["gateways", "APIKEY_PROVIDERS_GATEWAYS"],
    ["frontier-labs", "APIKEY_PROVIDERS_FRONTIER"],
@@ -60,7 +61,7 @@ test("APIKEY_PROVIDERS merges the 6 family files into 166 entries (no loss / no
      seen.add(k);
    }
  }
-  assert.equal(famTotal, 166, "families must partition all 166 providers");
+  assert.equal(famTotal, 167, "families must partition all 167 providers");
 });

 test("AI_PROVIDERS Proxy aggregates all sections; lookups resolve", () => {