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

feat: add ZenMux provider (Phase 2B of #3368) (#3429)

Paijoea0c0d8499bf
diff --git a/open-sse/config/providerRegistry.ts b/open-sse/config/providerRegistry.ts
index b441fd8251..41eeaebfe0 100644
--- a/open-sse/config/providerRegistry.ts
+++ b/open-sse/config/providerRegistry.ts
@@ -4318,6 +4318,92 @@ const _REGISTRY_EAGER: Record<string, RegistryEntry> = {
],
},
+ zenmux: {
+ id: "zenmux",
+ alias: "zm",
+ format: "openai",
+ executor: "default",
+ baseUrl: "https://zenmux.ai/api/v1/chat/completions",
+ modelsUrl: "https://zenmux.ai/api/v1/models",
+ authType: "apikey",
+ authHeader: "bearer",
+ defaultContextLength: 128000,
+ models: [
+ {
+ id: "google/gemini-3.1-pro-preview",
+ name: "Gemini 3.1 Pro Preview (ZenMux)",
+ contextLength: 1048576,
+ supportsVision: true,
+ toolCalling: true,
+ supportsReasoning: true,
+ },
+ {
+ id: "google/gemini-3-flash-preview",
+ name: "Gemini 3 Flash Preview (ZenMux)",
+ contextLength: 1048576,
+ supportsVision: true,
+ toolCalling: true,
+ supportsReasoning: true,
+ },
+ {
+ id: "openai/gpt-5",
+ name: "GPT-5 (ZenMux)",
+ contextLength: 400000,
+ supportsVision: true,
+ toolCalling: true,
+ supportsReasoning: true,
+ },
+ {
+ id: "anthropic/claude-sonnet-4.5",
+ name: "Claude Sonnet 4.5 (ZenMux)",
+ contextLength: 200000,
+ supportsVision: true,
+ toolCalling: true,
+ supportsReasoning: true,
+ },
+ {
+ id: "anthropic/claude-opus-4.5",
+ name: "Claude Opus 4.5 (ZenMux)",
+ contextLength: 200000,
+ supportsVision: true,
+ toolCalling: true,
+ supportsReasoning: true,
+ },
+ {
+ id: "deepseek/deepseek-chat",
+ name: "DeepSeek V3.2 Chat (ZenMux)",
+ contextLength: 128000,
+ supportsVision: false,
+ toolCalling: true,
+ supportsReasoning: false,
+ },
+ {
+ id: "x-ai/grok-4.1-fast",
+ name: "Grok 4.1 Fast (ZenMux)",
+ contextLength: 131072,
+ supportsVision: false,
+ toolCalling: true,
+ supportsReasoning: true,
+ },
+ {
+ id: "mistralai/mistral-large-2512",
+ name: "Mistral Large 2512 (ZenMux)",
+ contextLength: 128000,
+ supportsVision: true,
+ toolCalling: true,
+ supportsReasoning: false,
+ },
+ {
+ id: "z-ai/glm-4.6v-flash",
+ name: "GLM 4.6V Flash (ZenMux)",
+ contextLength: 128000,
+ supportsVision: true,
+ toolCalling: true,
+ supportsReasoning: false,
+ },
+ ],
+ },
+
theoldllm: {
id: "theoldllm",
alias: "tllm",
diff --git a/src/shared/constants/config.ts b/src/shared/constants/config.ts
index 85a93070a1..32a86de86d 100644
--- a/src/shared/constants/config.ts
+++ b/src/shared/constants/config.ts
@@ -29,6 +29,7 @@ export const PROVIDER_ENDPOINTS = {
minimax: "https://api.minimax.io/anthropic/v1/messages",
"minimax-cn": "https://api.minimaxi.com/anthropic/v1/messages",
crof: "https://crof.ai/v1/chat/completions",
+ zenmux: "https://zenmux.ai/api/v1/chat/completions",
openai: "https://api.openai.com/v1/chat/completions",
anthropic: "https://api.anthropic.com/v1/messages",
gemini: "https://generativelanguage.googleapis.com/v1beta/models",
diff --git a/src/shared/constants/providers.ts b/src/shared/constants/providers.ts
index 67df2b560a..de54566eee 100644
--- a/src/shared/constants/providers.ts
+++ b/src/shared/constants/providers.ts
@@ -2284,6 +2284,22 @@ export const APIKEY_PROVIDERS = {
apiHint:
"Discounted API proxy for 40+ models including GPT-5, Claude Opus 4.6, Claude Sonnet 4.6, Qwen 3.5. Get your API key at https://freeaiapikey.com/dashboard. Base URL: https://freeaiapikey.com/v1.",
},
+ zenmux: {
+ id: "zenmux",
+ alias: "zm",
+ name: "ZenMux",
+ icon: "neurology",
+ color: "#7C3AED",
+ textIcon: "ZM",
+ website: "https://zenmux.ai",
+ hasFree: true,
+ freeNote:
+ "Free tier includes access to Gemini 3 Flash, DeepSeek V3.2, Grok 4.1 Fast, Mistral Large, and more. Get your API key at https://zenmux.ai.",
+ authHint:
+ "Use your ZenMux API key in Authorization: Bearer <key>. ZenMux is fully OpenAI-compatible. Base URL: https://zenmux.ai/api/v1.",
+ apiHint:
+ "ZenMux exposes an OpenAI-compatible chat completions endpoint at /api/v1/chat/completions, plus Anthropic Messages (/api/anthropic/v1/messages) and Google Gemini (/api/vertex-ai) protocol surfaces. OmniRoute uses the OpenAI protocol.",
+ },
};
// Sub-categories within APIKEY_PROVIDERS (used by dashboard and catalog views).
diff --git a/tests/unit/zenmux-provider.test.ts b/tests/unit/zenmux-provider.test.ts
new file mode 100644
index 0000000000..1fd54b9320
--- /dev/null
+++ b/tests/unit/zenmux-provider.test.ts
@@ -0,0 +1,67 @@
+import test from "node:test";
+import assert from "node:assert/strict";
+
+const { APIKEY_PROVIDERS } = await import("../../src/shared/constants/providers.ts");
+const { PROVIDER_ENDPOINTS } = await import("../../src/shared/constants/config.ts");
+const { REGISTRY: providerRegistry } = await import("../../open-sse/config/providerRegistry.ts");
+
+test("ZenMux is registered as an API-key provider with the canonical identity", () => {
+ const zenmux = APIKEY_PROVIDERS.zenmux;
+ assert.ok(zenmux, "APIKEY_PROVIDERS.zenmux must be defined");
+ assert.equal(zenmux.id, "zenmux");
+ assert.equal(zenmux.alias, "zm");
+ assert.equal(zenmux.name, "ZenMux");
+ assert.equal(zenmux.website, "https://zenmux.ai");
+ assert.equal(typeof zenmux.textIcon, "string");
+ assert.equal(zenmux.hasFree, true);
+});
+
+test("ZenMux exposes the OpenAI-compatible chat completions URL", () => {
+ assert.equal(
+ PROVIDER_ENDPOINTS.zenmux,
+ "https://zenmux.ai/api/v1/chat/completions"
+ );
+});
+
+test("ZenMux registry entry uses OpenAI format with bearer apikey auth", () => {
+ const entry = providerRegistry.zenmux;
+ assert.ok(entry, "providerRegistry.zenmux must be defined");
+ assert.equal(entry.id, "zenmux");
+ assert.equal(entry.alias, "zm");
+ assert.equal(entry.format, "openai");
+ assert.equal(entry.executor, "default");
+ assert.equal(entry.authType, "apikey");
+ assert.equal(entry.authHeader, "bearer");
+ assert.equal(entry.baseUrl, "https://zenmux.ai/api/v1/chat/completions");
+ assert.equal(entry.modelsUrl, "https://zenmux.ai/api/v1/models");
+});
+
+test("ZenMux seed model list includes the headline families and unique ids", () => {
+ const models = providerRegistry.zenmux.models;
+ const ids = models.map((m: { id: string }) => m.id);
+ assert.ok(ids.length >= 8, "expect a non-trivial seed list of 8+ models");
+ assert.equal(new Set(ids).size, ids.length, "model ids must be unique");
+ // Provider-prefixed family coverage matching ZenMux's multi-protocol router
+ for (const family of [
+ "google/gemini-3",
+ "openai/gpt-5",
+ "anthropic/claude-sonnet",
+ "deepseek/deepseek",
+ "x-ai/grok-",
+ "mistralai/mistral",
+ ]) {
+ assert.ok(
+ ids.some((id: string) => id.startsWith(family)),
+ `seed list must include ${family}* model`
+ );
+ }
+});
+
+test("ZenMux models use provider/model naming convention", () => {
+ for (const model of providerRegistry.zenmux.models) {
+ assert.ok(
+ model.id.includes("/"),
+ `model id "${model.id}" must follow provider/model format (e.g. google/gemini-3.1-pro-preview)`
+ );
+ }
+});