Skip to content

Never stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 900+ contributors

fix(providers): support local OpenAI-style endpoints without auth

diegosouzapw968cbfeb1569
diff --git a/open-sse/config/providerRegistry.ts b/open-sse/config/providerRegistry.ts
index 7e3ef2080b..3d6c5caf5d 100644
--- a/open-sse/config/providerRegistry.ts
+++ b/open-sse/config/providerRegistry.ts
@@ -1248,6 +1248,73 @@ export const REGISTRY: Record<string, RegistryEntry> = {
authType: "apikey", // Some setups may use proxy with api key
authHeader: "bearer",
models: [], // Usually dynamic based on local models downloaded
+ passthroughModels: true,
+ },
+ vllm: {
+ id: "vllm",
+ alias: "vllm",
+ format: "openai",
+ executor: "default",
+ baseUrl: "http://localhost:8000/v1",
+ authType: "apikey",
+ authHeader: "bearer",
+ models: [],
+ passthroughModels: true,
+ },
+ llamafile: {
+ id: "llamafile",
+ alias: "llamafile",
+ format: "openai",
+ executor: "default",
+ baseUrl: "http://127.0.0.1:8080/v1",
+ authType: "apikey",
+ authHeader: "bearer",
+ models: [],
+ passthroughModels: true,
+ },
+ triton: {
+ id: "triton",
+ alias: "triton",
+ format: "openai",
+ executor: "default",
+ baseUrl: "http://localhost:8000/v1",
+ authType: "apikey",
+ authHeader: "bearer",
+ models: [],
+ passthroughModels: true,
+ },
+ "docker-model-runner": {
+ id: "docker-model-runner",
+ alias: "dmr",
+ format: "openai",
+ executor: "default",
+ baseUrl: "http://localhost:12434/v1",
+ authType: "apikey",
+ authHeader: "bearer",
+ models: [],
+ passthroughModels: true,
+ },
+ xinference: {
+ id: "xinference",
+ alias: "xinference",
+ format: "openai",
+ executor: "default",
+ baseUrl: "http://localhost:9997/v1",
+ authType: "apikey",
+ authHeader: "bearer",
+ models: [],
+ passthroughModels: true,
+ },
+ oobabooga: {
+ id: "oobabooga",
+ alias: "ooba",
+ format: "openai",
+ executor: "default",
+ baseUrl: "http://localhost:5000/v1",
+ authType: "apikey",
+ authHeader: "bearer",
+ models: [],
+ passthroughModels: true,
},
"ollama-cloud": {
diff --git a/open-sse/executors/default.ts b/open-sse/executors/default.ts
index 25327f7518..2ff1d9e3d2 100644
--- a/open-sse/executors/default.ts
+++ b/open-sse/executors/default.ts
@@ -51,6 +51,18 @@ function normalizeGigachatChatUrl(baseUrl) {
return `${normalized}/chat/completions`;
}
+function normalizeOpenAIChatUrl(baseUrl) {
+ const normalized = normalizeBaseUrl(baseUrl);
+ if (
+ normalized.endsWith("/chat/completions") ||
+ normalized.endsWith("/responses") ||
+ normalized.endsWith("/chat")
+ ) {
+ return normalized;
+ }
+ return normalized.endsWith("/v1") ? `${normalized}/chat/completions` : normalized;
+}
+
export class DefaultExecutor extends BaseExecutor {
constructor(provider) {
super(provider, PROVIDERS[provider] || PROVIDERS.openai);
@@ -110,6 +122,16 @@ export class DefaultExecutor extends BaseExecutor {
const baseUrl = credentials?.providerSpecificData?.baseUrl || this.config.baseUrl;
return normalizeGigachatChatUrl(baseUrl);
}
+ case "lm-studio":
+ case "vllm":
+ case "llamafile":
+ case "triton":
+ case "docker-model-runner":
+ case "xinference":
+ case "oobabooga": {
+ const baseUrl = credentials?.providerSpecificData?.baseUrl || this.config.baseUrl;
+ return normalizeOpenAIChatUrl(baseUrl);
+ }
case "claude":
case "glm":
case "glmt":
@@ -189,7 +211,10 @@ export class DefaultExecutor extends BaseExecutor {
headers["anthropic-version"] = "2023-06-01";
}
} else {
- headers["Authorization"] = `Bearer ${effectiveKey || credentials.accessToken}`;
+ const bearerToken = effectiveKey || credentials.accessToken;
+ if (bearerToken) {
+ headers["Authorization"] = `Bearer ${bearerToken}`;
+ }
}
}
diff --git a/src/app/(dashboard)/dashboard/providers/[id]/page.tsx b/src/app/(dashboard)/dashboard/providers/[id]/page.tsx
index 2c1940090b..988c21933f 100644
--- a/src/app/(dashboard)/dashboard/providers/[id]/page.tsx
+++ b/src/app/(dashboard)/dashboard/providers/[id]/page.tsx
@@ -23,10 +23,12 @@ import {
ProxyConfigModal,
} from "@/shared/components";
import {
+ LOCAL_PROVIDERS,
getProviderAlias,
isOpenAICompatibleProvider,
isAnthropicCompatibleProvider,
isClaudeCodeCompatibleProvider,
+ isSelfHostedChatProvider,
supportsApiKeyOnFreeProvider,
} from "@/shared/constants/providers";
import { getModelsByProviderId } from "@/shared/constants/models";
@@ -89,6 +91,11 @@ type CompatModelRow = {
};
type CompatModelMap = Map<string, CompatModelRow>;
+type LocalProviderMetadata = {
+ name?: string;
+ localDefault?: string;
+ [key: string]: unknown;
+};
function buildCompatMap(rows: CompatModelRow[]): CompatModelMap {
const m = new Map<string, CompatModelRow>();
@@ -5338,11 +5345,37 @@ const DEFAULT_PROVIDER_BASE_URLS: Record<string, string> = {
"searxng-search": "http://localhost:8888/search",
};
+function getLocalProviderMetadata(providerId?: string | null) {
+ if (!providerId || !isSelfHostedChatProvider(providerId)) return null;
+ return (LOCAL_PROVIDERS as Record<string, LocalProviderMetadata>)[providerId] || null;
+}
+
+function isBaseUrlConfigurableProvider(providerId?: string | null) {
+ return Boolean(
+ providerId &&
+ (CONFIGURABLE_BASE_URL_PROVIDERS.has(providerId) || isSelfHostedChatProvider(providerId))
+ );
+}
+
function getProviderBaseUrlDefault(providerId?: string | null) {
+ const localProvider = getLocalProviderMetadata(providerId);
+ if (typeof localProvider?.localDefault === "string" && localProvider.localDefault.trim()) {
+ return localProvider.localDefault;
+ }
return providerId ? DEFAULT_PROVIDER_BASE_URLS[providerId] || "" : "";
}
-function getProviderBaseUrlHint(providerId?: string | null, t?: (key: string) => string) {
+function getProviderBaseUrlHint(
+ providerId?: string | null,
+ t?: ((key: string, values?: Record<string, unknown>) => string) | null
+) {
+ const localProvider = getLocalProviderMetadata(providerId);
+ if (localProvider && t) {
+ return t("localProviderBaseUrlHint", {
+ provider: localProvider.name || providerId,
+ baseUrl: getProviderBaseUrlDefault(providerId),
+ });
+ }
switch (providerId) {
case "azure-openai":
return t ? t("azureOpenAiBaseUrlHint") : undefined;
@@ -5364,6 +5397,9 @@ function getProviderBaseUrlHint(providerId?: string | null, t?: (key: string) =>
}
function getProviderBaseUrlPlaceholder(providerId?: string | null) {
+ if (isSelfHostedChatProvider(providerId || "")) {
+ return getProviderBaseUrlDefault(providerId);
+ }
switch (providerId) {
case "azure-openai":
return "https://my-resource.openai.azure.com";
@@ -5434,13 +5470,15 @@ function AddApiKeyModal({
onClose,
}: AddApiKeyModalProps) {
const t = useTranslations("providers");
- const usesBaseUrl = CONFIGURABLE_BASE_URL_PROVIDERS.has(provider || "");
+ const usesBaseUrl = isBaseUrlConfigurableProvider(provider);
const defaultBaseUrl = getProviderBaseUrlDefault(provider);
const isVertex = provider === "vertex" || provider === "vertex-partner";
const defaultRegion = "us-central1";
const isGlm = provider === "glm" || provider === "glmt";
const isQoder = provider === "qoder";
const isCloudflare = provider === "cloudflare-ai";
+ const localProviderMetadata = getLocalProviderMetadata(provider);
+ const isLocalSelfHostedProvider = !!localProviderMetadata;
const isSearxng = provider === "searxng-search";
const isGooglePse = provider === "google-pse-search";
const isGrokWeb = provider === "grok-web";
@@ -5448,7 +5486,7 @@ function AddApiKeyModal({
const isBlackboxWeb = provider === "blackbox-web";
const isMuseSparkWeb = provider === "muse-spark-web";
const isWebSessionProvider = isGrokWeb || isPerplexityWeb || isBlackboxWeb || isMuseSparkWeb;
- const apiKeyOptional = isSearxng;
+ const apiKeyOptional = isSearxng || isLocalSelfHostedProvider;
const [formData, setFormData] = useState({
name: "",
@@ -5476,7 +5514,7 @@ function AddApiKeyModal({
? t("personalAccessTokenLabel")
: isWebSessionProvider
? t("sessionCookieLabel")
- : isSearxng
+ : apiKeyOptional
? `${t("apiKeyLabel")} (${t("optional").toLowerCase()})`
: t("apiKeyLabel");
const apiCredentialPlaceholder = isVertex
@@ -5491,7 +5529,7 @@ function AddApiKeyModal({
? t("museSparkWebCookiePlaceholder")
: isQoder
? t("qoderPatPlaceholder")
- : isSearxng
+ : apiKeyOptional
? t("optional")
: undefined;
const apiCredentialHint = isQoder
@@ -5504,9 +5542,13 @@ function AddApiKeyModal({
? t("blackboxWebCookieHint")
: isMuseSparkWeb
? t("museSparkWebCookieHint")
- : isSearxng
- ? t("apiKeyOptionalHint")
- : undefined;
+ : isLocalSelfHostedProvider
+ ? t("localProviderApiKeyOptionalHint", {
+ provider: localProviderMetadata?.name || providerName || provider || "",
+ })
+ : isSearxng
+ ? t("apiKeyOptionalHint")
+ : undefined;
const handleValidate = async () => {
setValidating(true);
@@ -5922,18 +5964,27 @@ function EditConnectionModal({ isOpen, connection, onSave, onClose }: EditConnec
const { emailsVisible: showEmail, toggleEmailVisibility: toggleShowEmail } =
useEmailPrivacyStore();
- const usesBaseUrl = CONFIGURABLE_BASE_URL_PROVIDERS.has(connection?.provider || "");
+ const usesBaseUrl = isBaseUrlConfigurableProvider(connection?.provider);
const defaultBaseUrl = getProviderBaseUrlDefault(connection?.provider);
const isVertex = connection?.provider === "vertex" || connection?.provider === "vertex-partner";
const isGlm = connection?.provider === "glm" || connection?.provider === "glmt";
const isCloudflare = connection?.provider === "cloudflare-ai";
const isCodex = connection?.provider === "codex";
const isClaude = connection?.provider === "claude";
+ const localProviderMetadata = getLocalProviderMetadata(connection?.provider);
+ const isLocalSelfHostedProvider = !!localProviderMetadata;
const isSearxng = connection?.provider === "searxng-search";
const isGooglePse = connection?.provider === "google-pse-search";
- const apiKeyOptional = isSearxng;
+ const apiKeyOptional = isSearxng || isLocalSelfHostedProvider;
const isCcCompatible = isClaudeCodeCompatibleProvider(connection?.provider);
const defaultRegion = "us-central1";
+ const apiCredentialHint = isLocalSelfHostedProvider
+ ? t("localProviderApiKeyOptionalHint", {
+ provider: localProviderMetadata?.name || connection?.provider || "",
+ })
+ : isSearxng
+ ? t("apiKeyOptionalHint")
+ : t("leaveBlankKeepCurrentApiKey");
useEffect(() => {
if (connection) {
@@ -6361,12 +6412,12 @@ function EditConnectionModal({ isOpen, connection, onSave, onClose }: EditConnec
<>
<div className="flex gap-2">
<Input
- label={isSearxng ? t("apiKeyOptionalLabel") : t("apiKeyLabel")}
+ label={apiKeyOptional ? t("apiKeyOptionalLabel") : t("apiKeyLabel")}
type="password"
value={formData.apiKey}
onChange={(e) => setFormData({ ...formData, apiKey: e.target.value })}
placeholder={isVertex ? t("vertexServiceAccountPlaceholder") : t("enterNewApiKey")}
- hint={isSearxng ? t("apiKeyOptionalHint") : t("leaveBlankKeepCurrentApiKey")}
+ hint={apiCredentialHint}
className="flex-1"
/>
<div className="pt-6">
diff --git a/src/app/api/providers/[id]/models/route.ts b/src/app/api/providers/[id]/models/route.ts
index 12ba06e398..700902ab1f 100755
--- a/src/app/api/providers/[id]/models/route.ts
+++ b/src/app/api/providers/[id]/models/route.ts
@@ -1,9 +1,11 @@
import { NextResponse } from "next/server";
import {
isClaudeCodeCompatibleProvider,
- isOpenAICompatibleProvider,
isAnthropicCompatibleProvider,
+ isOpenAICompatibleProvider,
+ isSelfHostedChatProvider,
} from "@/shared/constants/providers";
+import { getRegistryEntry } from "@omniroute/open-sse/config/providerRegistry.ts";
import { PROVIDER_MODELS } from "@/shared/constants/models";
import {
getProviderConnectionById,
@@ -51,6 +53,17 @@ function getProviderBaseUrl(providerSpecificData: unknown): string | null {
return typeof baseUrl === "string" && baseUrl.trim().length > 0 ? baseUrl : null;
}
+function isLocalOpenAIStyleProvider(provider: string): boolean {
+ return isSelfHostedChatProvider(provider);
+}
+
+function buildOptionalBearerHeaders(token: string | null | undefined): Record<string, string> {
+ return {
+ "Content-Type": "application/json",
+ ...(token ? { Authorization: `Bearer ${token}` } : {}),
+ };
+}
+
function normalizeAntigravityModelsResponse(data: unknown): Array<{ id: string; name: string }> {
const payload = asRecord(data).models;
@@ -610,14 +623,19 @@ export async function GET(
});
};
- if (isOpenAICompatibleProvider(provider)) {
+ if (isOpenAICompatibleProvider(provider) || isLocalOpenAIStyleProvider(provider)) {
const cachedResponse = maybeReturnCachedDiscovery();
if (cachedResponse) return cachedResponse;
const autoFetchDisabledResponse = maybeReturnAutoFetchDisabled();
if (autoFetchDisabledResponse) return autoFetchDisabledResponse;
- const baseUrl = getProviderBaseUrl(connection.providerSpecificData);
+ const registryEntry = isLocalOpenAIStyleProvider(provider)
+ ? getRegistryEntry(provider)
+ : null;
+ const baseUrl =
+ getProviderBaseUrl(connection.providerSpecificData) ||
+ (typeof registryEntry?.baseUrl === "string" ? registryEntry.baseUrl : null);
if (!baseUrl) {
const fallback = buildDiscoveryFallbackResponse({
cacheWarning: "Base URL unavailable — using cached catalog",
@@ -625,7 +643,11 @@ export async function GET(
});
if (fallback) return fallback;
return NextResponse.json(
- { error: "No base URL configured for OpenAI compatible provider" },
+ {
+ error: isOpenAICompatibleProvider(provider)
+ ? "No base URL configured for OpenAI compatible provider"
+ : "No base URL configured for local provider",
+ },
{ status: 400 }
);
}
@@ -650,6 +672,7 @@ export async function GET(
const uniqueEndpoints = [...new Set(endpoints)];
let models = null;
let lastErrorStatus = null;
+ const token = apiKey || accessToken;
for (const modelsUrl of uniqueEndpoints) {
try {
@@ -658,10 +681,7 @@ export async function GET(
guard: getProviderOutboundGuard(),
proxyConfig: proxy,
method: "GET",
- headers: {
- "Content-Type": "application/json",
- Authorization: `Bearer ${apiKey}`,
- },
+ headers: buildOptionalBearerHeaders(token),
});
if (response.ok) {
diff --git a/src/i18n/messages/en.json b/src/i18n/messages/en.json
index 876c894ce8..a6e8c99fdd 100644
--- a/src/i18n/messages/en.json
+++ b/src/i18n/messages/en.json
@@ -2149,6 +2149,7 @@
"museSparkWebCookieHint": "Paste the abra_sess cookie from meta.ai. A full cookie header also works.",
"apiKeyOptionalLabel": "API Key (optional)",
"apiKeyOptionalHint": "Optional. Leave blank if your SearXNG instance does not require authentication.",
+ "localProviderApiKeyOptionalHint": "Optional. Leave blank if your {provider} endpoint does not require authentication.",
"ccCompatibleNamePlaceholder": "CC Compatible Production",
"ccCompatibleNameHint": "Display name for this provider.",
"ccCompatiblePrefixPlaceholder": "cc",
@@ -2164,7 +2165,8 @@
"herokuBaseUrlHint": "Required: paste the Heroku Inference base URL. The app will append /v1/chat/completions.",
"databricksBaseUrlHint": "Required: paste the Databricks serving-endpoints base URL. The app will append /chat/completions.",
"snowflakeBaseUrlHint": "Required: paste the Snowflake account base URL. The app will append /api/v2/cortex/inference:complete.",
- "searxngBaseUrlHint": "Required: paste your SearXNG instance base URL. The app will use /search and request format=json. Local/private URLs require OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS=true for dashboard validation."
+ "searxngBaseUrlHint": "Required: paste your SearXNG instance base URL. The app will use /search and request format=json. Local/private URLs require OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS=true for dashboard validation.",
+ "localProviderBaseUrlHint": "Required: paste your {provider} OpenAI-compatible /v1 base URL (default: {baseUrl}). Local/private URLs require OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS=true for dashboard validation."
},
"settings": {
"title": "Settings",
diff --git a/src/i18n/messages/pt-BR.json b/src/i18n/messages/pt-BR.json
index b959961274..fac3e3cb86 100644
--- a/src/i18n/messages/pt-BR.json
+++ b/src/i18n/messages/pt-BR.json
@@ -2150,6 +2150,7 @@
"museSparkWebCookieHint": "Cole o cookie abra_sess do meta.ai. Um cabeçalho de cookie completo também funciona.",
"apiKeyOptionalLabel": "Chave de API (opcional)",
"apiKeyOptionalHint": "Opcional. Deixe em branco se sua instância SearXNG não exigir autenticação.",
+ "localProviderApiKeyOptionalHint": "Opcional. Deixe em branco se o endpoint {provider} não exigir autenticação.",
"ccCompatibleNamePlaceholder": "Produção CC Compatível",
"ccCompatibleNameHint": "Nome de exibição para este provedor.",
"ccCompatiblePrefixPlaceholder": "cc",
@@ -2165,7 +2166,8 @@
"herokuBaseUrlHint": "Obrigatório: cole a URL base do Heroku Inference. O app adicionará /v1/chat/completions.",
"databricksBaseUrlHint": "Obrigatório: cole a URL base do serving-endpoints do Databricks. O app adicionará /chat/completions.",
"snowflakeBaseUrlHint": "Obrigatório: cole a URL base da conta Snowflake. O app adicionará /api/v2/cortex/inference:complete.",
- "searxngBaseUrlHint": "Obrigatório: cole a URL base da sua instância SearXNG. O app usará /search e request format=json. URLs locais/privadas exigem OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS=true para validação no dashboard."
+ "searxngBaseUrlHint": "Obrigatório: cole a URL base da sua instância SearXNG. O app usará /search e request format=json. URLs locais/privadas exigem OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS=true para validação no dashboard.",
+ "localProviderBaseUrlHint": "Obrigatório: cole a URL base OpenAI-compatible /v1 do seu {provider} (padrão: {baseUrl}). URLs locais/privadas exigem OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS=true para validação no dashboard."
},
"settings": {
"title": "Configurações",
diff --git a/src/lib/providers/validation.ts b/src/lib/providers/validation.ts
index f9223ca263..dd038dacf2 100644
--- a/src/lib/providers/validation.ts
+++ b/src/lib/providers/validation.ts
@@ -13,6 +13,7 @@ import {
isClaudeCodeCompatibleProvider,
isAnthropicCompatibleProvider,
isOpenAICompatibleProvider,
+ isSelfHostedChatProvider,
} from "@/shared/constants/providers";
import {
SAFE_OUTBOUND_FETCH_PRESETS,
@@ -150,13 +151,13 @@ function withCustomUserAgent(init: RequestInit, providerSpecificData: any = {})
}
function buildBearerHeaders(apiKey: string, providerSpecificData: any = {}) {
- return applyCustomUserAgent(
- {
- "Content-Type": "application/json",
- Authorization: `Bearer ${apiKey}`,
- },
- providerSpecificData
- );
+ const headers: Record<string, string> = {
+ "Content-Type": "application/json",
+ };
+ if (apiKey) {
+ headers.Authorization = `Bearer ${apiKey}`;
+ }
+ return applyCustomUserAgent(headers, providerSpecificData);
}
async function validationRead(url: string, init: RequestInit) {
@@ -1646,7 +1647,7 @@ async function validateMuseSparkWebProvider({ apiKey, providerSpecificData = {}
}
export async function validateProviderApiKey({ provider, apiKey, providerSpecificData = {} }: any) {
- const requiresApiKey = provider !== "searxng-search";
+ const requiresApiKey = provider !== "searxng-search" && !isSelfHostedChatProvider(provider);
if (!provider || (requiresApiKey && !apiKey)) {
return { valid: false, error: "Provider and API key required", unsupported: false };
}
diff --git a/src/shared/constants/providers.ts b/src/shared/constants/providers.ts
index 17843fe5d4..97ddbb43b9 100644
--- a/src/shared/constants/providers.ts
+++ b/src/shared/constants/providers.ts
@@ -406,15 +406,6 @@ export const APIKEY_PROVIDERS = {
textIcon: "NB",
website: "https://nanobananaapi.ai",
},
- "lm-studio": {
- id: "lm-studio",
- alias: "lmstudio",
- name: "LM Studio",
- icon: "server",
- color: "#4A148C",
- textIcon: "LM",
- website: "https://lmstudio.ai",
- },
"ollama-cloud": {
id: "ollama-cloud",
alias: "ollamacloud",
@@ -962,6 +953,97 @@ export const APIKEY_PROVIDERS = {
// Local / Self-Hosted Providers
export const LOCAL_PROVIDERS = {
+ "lm-studio": {
+ id: "lm-studio",
+ alias: "lmstudio",
+ name: "LM Studio",
+ icon: "server",
+ color: "#4A148C",
+ textIcon: "LM",
+ website: "https://lmstudio.ai",
+ authHint:
+ "API key optional. Configure the local LM Studio OpenAI-compatible base URL (default: http://localhost:1234/v1).",
+ localDefault: "http://localhost:1234/v1",
+ passthroughModels: true,
+ },
+ vllm: {
+ id: "vllm",
+ alias: "vllm",
+ name: "vLLM",
+ icon: "memory",
+ color: "#0F766E",
+ textIcon: "VL",
+ website: "https://github.com/vllm-project/vllm",
+ authHint:
+ "API key optional. Configure the local vLLM OpenAI-compatible base URL (default: http://localhost:8000/v1).",
+ localDefault: "http://localhost:8000/v1",
+ passthroughModels: true,
+ },
+ llamafile: {
+ id: "llamafile",
+ alias: "llamafile",
+ name: "Llamafile",
+ icon: "article",
+ color: "#EA580C",
+ textIcon: "LF",
+ website: "https://github.com/Mozilla-Ocho/llamafile",
+ authHint:
+ "API key optional. Configure the local Llamafile OpenAI-compatible base URL (default: http://127.0.0.1:8080/v1).",
+ localDefault: "http://127.0.0.1:8080/v1",
+ passthroughModels: true,
+ },
+ triton: {
+ id: "triton",
+ alias: "triton",
+ name: "NVIDIA Triton",
+ icon: "developer_board",
+ color: "#76B900",
+ textIcon: "TR",
+ website: "https://developer.nvidia.com/triton-inference-server",
+ authHint:
+ "API key optional. Configure the Triton OpenAI-compatible base URL (default: http://localhost:8000/v1).",
+ localDefault: "http://localhost:8000/v1",
+ passthroughModels: true,
+ },
+ "docker-model-runner": {
+ id: "docker-model-runner",
+ alias: "dmr",
+ name: "Docker Model Runner",
+ icon: "inventory_2",
+ color: "#2496ED",
+ textIcon: "DM",
+ website: "https://docs.docker.com/ai/model-runner/",
+ authHint:
+ "API key optional. Configure the local Docker Model Runner OpenAI-compatible base URL (default: http://localhost:12434/v1).",
+ localDefault: "http://localhost:12434/v1",
+ passthroughModels: true,
+ },
+ xinference: {
+ id: "xinference",
+ alias: "xinference",
+ name: "XInference",
+ icon: "hub",
+ color: "#DC2626",
+ textIcon: "XI",
+ website: "https://inference.readthedocs.io",
+ authHint:
+ "API key optional. Configure the local XInference OpenAI-compatible base URL (default: http://localhost:9997/v1).",
+ localDefault: "http://localhost:9997/v1",
+ passthroughModels: true,
+ },
+ oobabooga: {
+ id: "oobabooga",
+ alias: "ooba",
+ name: "oobabooga",
+ icon: "dns",
+ color: "#8B5CF6",
+ textIcon: "OO",
+ website: "https://github.com/oobabooga/text-generation-webui",
+ authHint:
+ "API key optional. Configure the local oobabooga OpenAI-compatible base URL (default: http://localhost:5000/v1).",
+ localDefault: "http://localhost:5000/v1",
+ passthroughModels: true,
+ },
sdwebui: {
id: "sdwebui",
alias: "sdwebui",
@@ -1174,6 +1256,27 @@ export function isClaudeCodeCompatibleProvider(providerId) {
return typeof providerId === "string" && providerId.startsWith(CLAUDE_CODE_COMPATIBLE_PREFIX);
}
+export function isLocalProvider(providerId) {
+ return (
+ typeof providerId === "string" &&
+ Object.prototype.hasOwnProperty.call(LOCAL_PROVIDERS, providerId)
+ );
+}
+
+export const SELF_HOSTED_CHAT_PROVIDER_IDS = new Set([
+ "lm-studio",
+ "vllm",
+ "llamafile",
+ "triton",
+ "docker-model-runner",
+ "xinference",
+ "oobabooga",
+]);
+
+export function isSelfHostedChatProvider(providerId) {
+ return typeof providerId === "string" && SELF_HOSTED_CHAT_PROVIDER_IDS.has(providerId);
+}
+
// All providers (combined)
export const AI_PROVIDERS = {
...FREE_PROVIDERS,
diff --git a/src/shared/validation/schemas.ts b/src/shared/validation/schemas.ts
index 10205fcd9d..69da538de1 100644
--- a/src/shared/validation/schemas.ts
+++ b/src/shared/validation/schemas.ts
@@ -1,5 +1,6 @@
import { z } from "zod";
import { SUPPORTED_BATCH_ENDPOINTS } from "@/shared/constants/batchEndpoints";
+import { isLocalProvider } from "@/shared/constants/providers";
import { HIDEABLE_SIDEBAR_ITEM_IDS } from "@/shared/constants/sidebarVisibility";
import { isForbiddenUpstreamHeaderName } from "@/shared/constants/upstreamHeaders";
@@ -243,8 +244,8 @@ export const createProviderSchema = z
})
.superRefine((data, ctx) => {
const apiKey = typeof data.apiKey === "string" ? data.apiKey.trim() : "";
- const apiKeyOptionalProviders = new Set(["searxng-search", "sdwebui", "comfyui"]);
- if (!apiKeyOptionalProviders.has(data.provider) && apiKey.length === 0) {
+ const apiKeyOptional = data.provider === "searxng-search" || isLocalProvider(data.provider);
+ if (!apiKeyOptional && apiKey.length === 0) {
ctx.addIssue({
code: z.ZodIssueCode.custom,
message: "API key is required",
diff --git a/tests/unit/executor-default-base.test.ts b/tests/unit/executor-default-base.test.ts
index ce1d8af75a..40fbc1df16 100644
--- a/tests/unit/executor-default-base.test.ts
+++ b/tests/unit/executor-default-base.test.ts
@@ -245,6 +245,20 @@ test("DefaultExecutor.buildHeaders handles GLM, default auth and anthropic-compa
assert.equal(anthropicHeaders.Accept, "text/event-stream");
});
+test("DefaultExecutor local OpenAI-style providers honor custom base URLs and skip empty bearer headers", () => {
+ const lmStudio = new DefaultExecutor("lm-studio");
+ const vllm = new DefaultExecutor("vllm");
+
+ const lmStudioUrl = lmStudio.buildUrl("local-model", true, 0, {
+ providerSpecificData: { baseUrl: "http://127.0.0.1:4321/v1" },
+ });
+ const vllmHeaders = vllm.buildHeaders({}, false);
+
+ assert.equal(lmStudioUrl, "http://127.0.0.1:4321/v1/chat/completions");
+ assert.equal(vllmHeaders.Authorization, undefined);
+ assert.equal(vllmHeaders.Accept, "application/json");
+});
+
test("DefaultExecutor.buildHeaders handles Snowflake PATs and GigaChat access tokens", () => {
const snowflake = new DefaultExecutor("snowflake");
const gigachat = new DefaultExecutor("gigachat");
diff --git a/tests/unit/provider-models-route.test.ts b/tests/unit/provider-models-route.test.ts
index 2eb15c7303..0b8dd55896 100644
--- a/tests/unit/provider-models-route.test.ts
+++ b/tests/unit/provider-models-route.test.ts
@@ -208,6 +208,32 @@ test("provider models route returns static catalog entries for providers with ha
assert.equal(body.models.length, 8);
});
+test("provider models route discovers local OpenAI-style models without requiring an API key", async () => {
+ process.env.OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS = "true";
+
+ const connection = await seedConnection("lm-studio", {
+ providerSpecificData: {
+ baseUrl: "http://localhost:1234/v1",
+ },
+ });
+
+ globalThis.fetch = async (url, init = {}) => {
+ assert.equal(String(url), "http://localhost:1234/v1/models");
+ assert.equal(init.headers.Authorization, undefined);
+ return Response.json({
+ data: [{ id: "local-model", name: "Local Model" }],
+ });
+ };
+
+ const response = await callRoute(connection.id);
+ const body = (await response.json()) as any;
+
+ assert.equal(response.status, 200);
+ assert.equal(body.provider, "lm-studio");
+ assert.equal(body.source, "api");
+ assert.deepEqual(body.models, [{ id: "local-model", name: "Local Model" }]);
+});
+
test("provider models route returns the local catalog for built-in image providers", async () => {
const connection = await seedConnection("topaz", {
apiKey: "topaz-key",
diff --git a/tests/unit/provider-validation-specialty.test.ts b/tests/unit/provider-validation-specialty.test.ts
index 2a866a154e..6aab7389c6 100644
--- a/tests/unit/provider-validation-specialty.test.ts
+++ b/tests/unit/provider-validation-specialty.test.ts
@@ -364,6 +364,43 @@ test("google PSE validator requires cx", async () => {
assert.equal(result.error, "Programmable Search Engine ID (cx) is required");
});
+test("local OpenAI-style providers validate without sending Authorization when apiKey is blank", async () => {
+ const originalAllowPrivateProviderUrls = process.env.OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS;
+ process.env.OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS = "true";
+ const calls = [];
+
+ try {
+ globalThis.fetch = async (url, init = {}) => {
+ calls.push({ url: String(url), headers: init.headers || {} });
+ return new Response(JSON.stringify({ data: [{ id: "local-model" }] }), { status: 200 });
+ };
+
+ const lmStudio = await validateProviderApiKey({
+ provider: "lm-studio",
+ providerSpecificData: { baseUrl: "http://localhost:1234/v1" },
+ });
+ const vllm = await validateProviderApiKey({
+ provider: "vllm",
+ providerSpecificData: { baseUrl: "http://localhost:8000/v1" },
+ });
+
+ assert.equal(lmStudio.valid, true);
+ assert.equal(vllm.valid, true);
+ assert.deepEqual(
+ calls.map((call) => call.url),
+ ["http://localhost:1234/v1/models", "http://localhost:8000/v1/models"]
+ );
+ assert.equal(calls[0].headers.Authorization, undefined);
+ assert.equal(calls[1].headers.Authorization, undefined);
+ } finally {
+ if (originalAllowPrivateProviderUrls === undefined) {
+ delete process.env.OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS;
+ } else {
+ process.env.OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS = originalAllowPrivateProviderUrls;
+ }
+ }
+});
+
test("OpenAI-compatible validator covers /responses mode and final ping fallback", async () => {
const calls = [];
globalThis.fetch = async (url, init = {}) => {
diff --git a/tests/unit/providers-page-utils.test.ts b/tests/unit/providers-page-utils.test.ts
index e1320efd46..1dc7a4f43b 100644
--- a/tests/unit/providers-page-utils.test.ts
+++ b/tests/unit/providers-page-utils.test.ts
@@ -216,6 +216,7 @@ test("configured-only preference storage round-trips correctly", () => {
test("static catalog entries resolve local, search, audio, web-cookie and upstream providers", () => {
const localProvider = providerPageUtils.resolveDashboardProviderInfo("sdwebui");
+ const localChatProvider = providerPageUtils.resolveDashboardProviderInfo("lm-studio");
const searchProvider = providerPageUtils.resolveDashboardProviderInfo("brave-search");
const audioProvider = providerPageUtils.resolveDashboardProviderInfo("assemblyai");
const webCookieProvider = providerPageUtils.resolveDashboardProviderInfo("grok-web");
@@ -226,6 +227,8 @@ test("static catalog entries resolve local, search, audio, web-cookie and upstre
assert.equal(localProvider?.category, "local");
assert.equal(localProvider?.name, providers.LOCAL_PROVIDERS.sdwebui.name);
+ assert.equal(localChatProvider?.category, "local");
+ assert.equal(localChatProvider?.name, providers.LOCAL_PROVIDERS["lm-studio"].name);
assert.equal(searchProvider?.category, "search");
assert.equal(searchProvider?.name, providers.SEARCH_PROVIDERS["brave-search"].name);
@@ -255,6 +258,8 @@ test("static catalog entries resolve local, search, audio, web-cookie and upstre
test("managed provider connection ids include supported static categories and exclude upstream proxy", () => {
assert.equal(providerCatalog.isManagedProviderConnectionId("qoder"), true);
assert.equal(providerCatalog.isManagedProviderConnectionId("sdwebui"), true);
+ assert.equal(providerCatalog.isManagedProviderConnectionId("lm-studio"), true);
+ assert.equal(providerCatalog.isManagedProviderConnectionId("vllm"), true);
assert.equal(providerCatalog.isManagedProviderConnectionId("assemblyai"), true);
assert.equal(providerCatalog.isManagedProviderConnectionId("grok-web"), true);
assert.equal(providerCatalog.isManagedProviderConnectionId("perplexity-web"), true);
@@ -270,6 +275,10 @@ test("grok-web taxonomy stays web-cookie only and does not leak into api-key ent
assert.equal("grok-web" in providers.WEB_COOKIE_PROVIDERS, true);
assert.equal("sdwebui" in providers.APIKEY_PROVIDERS, false);
assert.equal("sdwebui" in providers.LOCAL_PROVIDERS, true);
+ assert.equal("lm-studio" in providers.APIKEY_PROVIDERS, false);
+ assert.equal("lm-studio" in providers.LOCAL_PROVIDERS, true);
+ assert.equal("vllm" in providers.APIKEY_PROVIDERS, false);
+ assert.equal("vllm" in providers.LOCAL_PROVIDERS, true);
assert.equal("comfyui" in providers.APIKEY_PROVIDERS, false);
assert.equal("comfyui" in providers.LOCAL_PROVIDERS, true);
assert.equal("blackbox-web" in providers.APIKEY_PROVIDERS, false);
@@ -295,6 +304,14 @@ test("grok-web taxonomy stays web-cookie only and does not leak into api-key ent
apiKeyEntries.some((entry) => entry.providerId === "comfyui"),
false
);
+ assert.equal(
+ apiKeyEntries.some((entry) => entry.providerId === "lm-studio"),
+ false
+ );
+ assert.equal(
+ apiKeyEntries.some((entry) => entry.providerId === "vllm"),
+ false
+ );
assert.equal(
apiKeyEntries.some((entry) => entry.providerId === "grok-web"),
false
@@ -307,6 +324,14 @@ test("grok-web taxonomy stays web-cookie only and does not leak into api-key ent
localEntries.some((entry) => entry.providerId === "comfyui"),
true
);
+ assert.equal(
+ localEntries.some((entry) => entry.providerId === "lm-studio"),
+ true
+ );
+ assert.equal(
+ localEntries.some((entry) => entry.providerId === "vllm"),
+ true
+ );
assert.equal(
webCookieEntries.some((entry) => entry.providerId === "grok-web"),
true
diff --git a/tests/unit/providers-route-managed-catalog.test.ts b/tests/unit/providers-route-managed-catalog.test.ts
index 09375c9cb8..2fd14f0488 100644
--- a/tests/unit/providers-route-managed-catalog.test.ts
+++ b/tests/unit/providers-route-managed-catalog.test.ts
@@ -40,6 +40,76 @@ test("providers route accepts managed local, audio, web-cookie and search provid
},
},
},
+ {
+ provider: "lm-studio",
+ body: {
+ provider: "lm-studio",
+ name: "LM Studio Local",
+ providerSpecificData: {
+ baseUrl: "http://localhost:1234/v1",
+ },
+ },
+ },
+ {
+ provider: "vllm",
+ body: {
+ provider: "vllm",
+ name: "vLLM Local",
+ providerSpecificData: {
+ baseUrl: "http://localhost:8000/v1",
+ },
+ },
+ },
+ {
+ provider: "llamafile",
+ body: {
+ provider: "llamafile",
+ name: "Llamafile Local",
+ providerSpecificData: {
+ baseUrl: "http://127.0.0.1:8080/v1",
+ },
+ },
+ },
+ {
+ provider: "triton",
+ body: {
+ provider: "triton",
+ name: "Triton Local",
+ providerSpecificData: {
+ baseUrl: "http://localhost:8000/v1",
+ },
+ },
+ },
+ {
+ provider: "docker-model-runner",
+ body: {
+ provider: "docker-model-runner",
+ name: "Docker Model Runner Local",
+ providerSpecificData: {
+ baseUrl: "http://localhost:12434/v1",
+ },
+ },
+ },
+ {
+ provider: "xinference",
+ body: {
+ provider: "xinference",
+ name: "XInference Local",
+ providerSpecificData: {
+ baseUrl: "http://localhost:9997/v1",
+ },
+ },
+ },
+ {
+ provider: "oobabooga",
+ body: {
+ provider: "oobabooga",
+ name: "oobabooga Local",
+ providerSpecificData: {
+ baseUrl: "http://localhost:5000/v1",
+ },
+ },
+ },
{
provider: "assemblyai",
body: {