feat(providers): add Venice AI provider

OpenAI-compatible apikey provider (chat/embedding/image) with dynamic
model discovery via modelsFetcher + passthroughModels. No executor needed.

Co-authored-by: Cursor <cursoragent@cursor.com>

open-sse/providers/registry/index.js

@@ -1,6 +1,6 @@
 // Auto-generated: static imports of all registry entries
-import p0 from "./alicode.js";
-import p1 from "./alicode-intl.js";
+import p0 from "./alicode-intl.js";
+import p1 from "./alicode.js";
 import p2 from "./anthropic.js";
 import p3 from "./antigravity.js";
 import p4 from "./assemblyai.js";
@@ -31,12 +31,12 @@ import p28 from "./exa.js";
 import p29 from "./fal-ai.js";
 import p30 from "./firecrawl.js";
 import p31 from "./fireworks.js";
-import p32 from "./gemini.js";
-import p33 from "./gemini-cli.js";
+import p32 from "./gemini-cli.js";
+import p33 from "./gemini.js";
 import p34 from "./github.js";
 import p35 from "./gitlab.js";
-import p36 from "./glm.js";
-import p37 from "./glm-cn.js";
+import p36 from "./glm-cn.js";
+import p37 from "./glm.js";
 import p38 from "./google-pse.js";
 import p39 from "./google-tts.js";
 import p40 from "./grok-web.js";
@@ -48,27 +48,27 @@ import p45 from "./inworld.js";
 import p46 from "./jina-ai.js";
 import p47 from "./jina-reader.js";
 import p48 from "./kilocode.js";
-import p49 from "./kimi.js";
-import p50 from "./kimi-coding.js";
+import p49 from "./kimi-coding.js";
+import p50 from "./kimi.js";
 import p51 from "./kiro.js";
 import p52 from "./linkup.js";
 import p53 from "./local-device.js";
 import p54 from "./mimo-free.js";
-import p55 from "./minimax.js";
-import p56 from "./minimax-cn.js";
+import p55 from "./minimax-cn.js";
+import p56 from "./minimax.js";
 import p57 from "./mistral.js";
 import p58 from "./mmf.js";
 import p59 from "./nanobanana.js";
 import p60 from "./nebius.js";
 import p61 from "./nvidia.js";
-import p62 from "./ollama.js";
-import p63 from "./ollama-local.js";
+import p62 from "./ollama-local.js";
+import p63 from "./ollama.js";
 import p64 from "./openai.js";
-import p65 from "./opencode.js";
-import p66 from "./opencode-go.js";
+import p65 from "./opencode-go.js";
+import p66 from "./opencode.js";
 import p67 from "./openrouter.js";
-import p68 from "./perplexity.js";
-import p69 from "./perplexity-web.js";
+import p68 from "./perplexity-web.js";
+import p69 from "./perplexity.js";
 import p70 from "./playht.js";
 import p71 from "./qoder.js";
 import p72 from "./qwen.js";
@@ -84,15 +84,16 @@ import p81 from "./tavily.js";
 import p82 from "./together.js";
 import p83 from "./topaz.js";
 import p84 from "./tortoise.js";
-import p85 from "./vercel-ai-gateway.js";
-import p86 from "./vertex.js";
+import p85 from "./venice.js";
+import p86 from "./vercel-ai-gateway.js";
 import p87 from "./vertex-partner.js";
-import p88 from "./volcengine-ark.js";
-import p89 from "./voyage-ai.js";
-import p90 from "./xai.js";
-import p91 from "./xiaomi-mimo.js";
-import p92 from "./xiaomi-tokenplan.js";
-import p93 from "./youcom.js";
+import p88 from "./vertex.js";
+import p89 from "./volcengine-ark.js";
+import p90 from "./voyage-ai.js";
+import p91 from "./xai.js";
+import p92 from "./xiaomi-mimo.js";
+import p93 from "./xiaomi-tokenplan.js";
+import p94 from "./youcom.js";
 
 export default [
   p0,
@@ -188,5 +189,6 @@ export default [
   p90,
   p91,
   p92,
-  p93
+  p93,
+  p94
 ];

open-sse/providers/registry/venice.js

@@ -0,0 +1,56 @@
+export default {
+  id: "venice",
+  priority: 115,
+  alias: "venice",
+  aliases: [
+    "vn",
+  ],
+  uiAlias: "venice",
+  display: {
+    name: "Venice AI",
+    icon: "shield",
+    color: "#DC2626",
+    textIcon: "VE",
+    website: "https://venice.ai",
+    notice: {
+      text: "OpenAI-compatible. Private inference + uncensored models (Venice Uncensored, GLM, Qwen, DeepSeek, Llama).",
+      apiKeyUrl: "https://venice.ai/settings/api",
+    },
+  },
+  category: "apikey",
+  transport: {
+    baseUrl: "https://api.venice.ai/api/v1/chat/completions",
+    validateUrl: "https://api.venice.ai/api/v1/models",
+    thinkingFormat: "openai",
+  },
+  // Curated seed; the full live catalogue (90+ text models) is fetched via
+  // modelsFetcher and any other id is accepted via passthroughModels.
+  models: [
+    { id: "venice-uncensored-1-2", name: "Venice Uncensored 1.2" },
+    { id: "zai-org-glm-5", name: "GLM-5" },
+    { id: "qwen3-235b-a22b-instruct-2507", name: "Qwen3 235B A22B Instruct" },
+    { id: "qwen3-coder-480b-a35b-instruct-turbo", name: "Qwen3 Coder 480B A35B Turbo" },
+    { id: "qwen3-vl-235b-a22b", name: "Qwen3 VL 235B A22B" },
+    { id: "deepseek-v4-pro", name: "DeepSeek V4 Pro" },
+    { id: "llama-3.3-70b", name: "Llama 3.3 70B" },
+    { id: "hermes-3-llama-3.1-405b", name: "Hermes 3 Llama 3.1 405B" },
+    { id: "mistral-small-3-2-24b-instruct", name: "Mistral Small 3.2 24B" },
+    { id: "text-embedding-3-large", name: "Text Embedding 3 Large", kind: "embedding" },
+    { id: "text-embedding-bge-m3", name: "BGE-M3 Embedding", kind: "embedding" },
+    { id: "text-embedding-qwen3-8b", name: "Qwen3 8B Embedding", kind: "embedding" },
+    { id: "venice-sd35", name: "Venice SD3.5", params: ["n", "size"], kind: "image" },
+    { id: "flux-2-pro", name: "FLUX.2 Pro", params: ["n", "size"], kind: "image" },
+    { id: "gpt-image-2", name: "GPT Image 2 (via Venice)", params: ["n", "size", "quality"], kind: "image" },
+  ],
+  serviceKinds: ["llm", "embedding", "image"],
+  embeddingConfig: {
+    baseUrl: "https://api.venice.ai/api/v1/embeddings",
+    authType: "apikey",
+    authHeader: "bearer",
+  },
+  imageConfig: {
+    baseUrl: "https://api.venice.ai/api/v1/images/generations",
+  },
+  modelsFetcher: { url: "https://api.venice.ai/api/v1/models", type: "openai" },
+  passthroughModels: true,
+};

tests/unit/venice-provider.test.js

@@ -0,0 +1,41 @@
+import { describe, expect, it } from "vitest";
+
+import REGISTRY from "../../open-sse/providers/registry/index.js";
+import { PROVIDERS, PROVIDER_MODELS } from "../../open-sse/providers/index.js";
+
+describe("Venice AI provider", () => {
+  const venice = REGISTRY.find((e) => e.id === "venice");
+
+  it("is registered as an OpenAI-compatible apikey provider", () => {
+    expect(venice).toBeDefined();
+    expect(venice.category).toBe("apikey");
+    expect(venice.transport.baseUrl).toBe("https://api.venice.ai/api/v1/chat/completions");
+    expect(venice.alias).toBe("venice");
+    expect(venice.aliases).toContain("vn");
+  });
+
+  it("enables dynamic model discovery and passthrough", () => {
+    expect(venice.passthroughModels).toBe(true);
+    expect(venice.modelsFetcher).toMatchObject({
+      url: "https://api.venice.ai/api/v1/models",
+      type: "openai",
+    });
+  });
+
+  it("builds into the runtime PROVIDERS map with the openai format default", () => {
+    expect(PROVIDERS.venice).toBeDefined();
+    expect(PROVIDERS.venice.format).toBe("openai");
+    expect(PROVIDERS.venice.baseUrl).toBe("https://api.venice.ai/api/v1/chat/completions");
+  });
+
+  it("exposes its seed models (incl. the signature uncensored model)", () => {
+    const ids = (PROVIDER_MODELS.venice || []).map((m) => m.id);
+    expect(ids.length).toBeGreaterThan(0);
+    expect(ids).toContain("venice-uncensored-1-2");
+  });
+
+  it("keeps every registry id unique after adding venice", () => {
+    const ids = REGISTRY.map((e) => e.id);
+    expect(new Set(ids).size).toBe(ids.length);
+  });
+});