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(providers): add Veo AI Free as web wrapper provider (#2366)

Paijodd24571949d5
diff --git a/open-sse/executors/index.ts b/open-sse/executors/index.ts
index 65466d62ee..b084fd2c8b 100644
--- a/open-sse/executors/index.ts
+++ b/open-sse/executors/index.ts
@@ -28,6 +28,7 @@ import { DevinCliExecutor } from "./devin-cli.ts";
import { DeepSeekWebExecutor } from "./deepseek-web.ts";
import { DeepSeekWebWithAutoRefreshExecutor } from "./deepseek-web-with-auto-refresh.ts";
import { CopilotWebExecutor } from "./copilot-web.ts";
+import { VeoAIFreeWebExecutor } from "./veoaifree-web.ts";
const executors = {
antigravity: new AntigravityExecutor(),
@@ -78,6 +79,8 @@ const executors = {
"ds-web": new DeepSeekWebWithAutoRefreshExecutor(), // Alias
"copilot-web": new CopilotWebExecutor(),
copilot: new CopilotWebExecutor(), // Alias
+ "veoaifree-web": new VeoAIFreeWebExecutor(),
+ "veo-free": new VeoAIFreeWebExecutor(), // Alias
};
const defaultCache = new Map();
@@ -122,5 +125,6 @@ export { PetalsExecutor } from "./petals.ts";
export { WindsurfExecutor } from "./windsurf.ts";
export { DevinCliExecutor } from "./devin-cli.ts";
export { CopilotWebExecutor } from "./copilot-web.ts";
+export { VeoAIFreeWebExecutor } from "./veoaifree-web.ts";
export { DeepSeekWebExecutor } from "./deepseek-web.ts";
export { DeepSeekWebWithAutoRefreshExecutor } from "./deepseek-web-with-auto-refresh.ts";
diff --git a/open-sse/executors/veoaifree-web.ts b/open-sse/executors/veoaifree-web.ts
new file mode 100644
index 0000000000..5f43b8abc1
--- /dev/null
+++ b/open-sse/executors/veoaifree-web.ts
@@ -0,0 +1,279 @@
+/**
+ * VeoAIFreeWebExecutor — Veo AI Free Multi-Tool Provider
+ *
+ * Routes requests through veoaifree.com's WordPress AJAX API.
+ * Supports: text-to-video, image-to-video, image generation, TTS, prompt enhancement.
+ *
+ * No auth required. Rate limited to 6 requests/hour per IP.
+ */
+import { BaseExecutor, type ExecuteInput } from "./base.ts";
+
+const BASE_URL = "https://veoaifree.com";
+const AJAX_URL = `${BASE_URL}/wp-admin/admin-ajax.php`;
+const TTS_URL = `${BASE_URL}/video/googletts.php`;
+const USER_AGENT =
+ "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/126.0.0.0 Safari/537.36";
+const POLL_INTERVAL_MS = 20_000;
+const MAX_POLLS = 30; // 10 minutes max
+
+// ─── Helpers ────────────────────────────────────────────────────────────────
+
+async function fetchNonce(): Promise<string> {
+ const res = await fetch(BASE_URL, { headers: { "User-Agent": USER_AGENT } });
+ const html = await res.text();
+ const match = html.match(/nonce":"([a-f0-9]+)"/);
+ if (!match) throw new Error("Failed to extract CSRF nonce from veoaifree.com");
+ return match[1];
+}
+
+async function postAjax(nonce: string, params: Record<string, string>): Promise<string> {
+ const body = new URLSearchParams({ action: "veo_video_generator", nonce, ...params });
+ const res = await fetch(AJAX_URL, {
+ method: "POST",
+ headers: {
+ "Content-Type": "application/x-www-form-urlencoded",
+ "User-Agent": USER_AGENT,
+ Origin: BASE_URL,
+ Referer: `${BASE_URL}/`,
+ },
+ body: body.toString(),
+ });
+ return res.text();
+}
+
+function jsonResp(data: unknown, status = 200): Response {
+ return new Response(JSON.stringify(data), {
+ status,
+ headers: { "Content-Type": "application/json" },
+ });
+}
+
+function errResp(message: string, status = 502): Response {
+ return jsonResp({ error: { message } }, status);
+}
+
+// ─── Intent Detection ───────────────────────────────────────────────────────
+
+type ToolIntent = "video" | "image" | "tts" | "enhance";
+
+export function detectIntent(model?: string, prompt?: string): ToolIntent {
+ const m = (model || "").toLowerCase();
+ if (m.includes("tts") || m.includes("speech") || m.includes("audio")) return "tts";
+ if (m.includes("image") || m.includes("banana") || m.includes("imagen")) return "image";
+ if (m.includes("enhance") || m.includes("prompt")) return "enhance";
+ if (m.includes("video") || m.includes("veo") || m.includes("seedance")) return "video";
+ // Auto-detect from prompt
+ const p = (prompt || "").toLowerCase();
+ if (p.startsWith("generate image") || p.startsWith("create image") || p.startsWith("draw "))
+ return "image";
+ if (p.startsWith("enhance") || p.startsWith("improve prompt")) return "enhance";
+ return "video"; // default
+}
+
+// ─── Tool Handlers ──────────────────────────────────────────────────────────
+
+async function handleVideo(nonce: string, prompt: string, aspectRatio: string): Promise<Response> {
+ // Generate
+ const genResult = await postAjax(nonce, {
+ prompt,
+ totalVariations: "1",
+ aspectRatio,
+ actionType: "full-video-generate",
+ });
+ const sceneData = genResult.trim();
+ if (!sceneData || sceneData === "0" || sceneData.toLowerCase().includes("error")) {
+ return errResp(`Video generation failed: ${sceneData}`);
+ }
+
+ // Poll
+ for (let i = 0; i < MAX_POLLS; i++) {
+ await new Promise((r) => setTimeout(r, POLL_INTERVAL_MS));
+ try {
+ const pollResult = await postAjax(nonce, {
+ sceneData,
+ actionType: "final-video-results",
+ });
+ const trimmed = pollResult.trim();
+ if (trimmed && trimmed !== "0" && !trimmed.toLowerCase().includes("error")) {
+ const urls = trimmed
+ .split(/[,\n]/)
+ .map((u) => u.trim())
+ .filter((u) => u.startsWith("http"));
+ if (urls.length > 0) {
+ return jsonResp({
+ object: "video.generation",
+ data: urls.map((url) => ({ url, type: "video" })),
+ status: "completed",
+ });
+ }
+ }
+ } catch {
+ /* continue polling */
+ }
+ }
+ return errResp("Video generation timed out after 10 minutes", 504);
+}
+
+async function handleImage(nonce: string, prompt: string, aspectRatio: string): Promise<Response> {
+ const result = await postAjax(nonce, {
+ promptIMG: prompt,
+ totalVariationsIMG: "1",
+ aspectRatioIMG: aspectRatio,
+ actionType: "banan-image-generator",
+ });
+ const trimmed = result.trim();
+ if (!trimmed || trimmed === "0" || trimmed.toLowerCase().includes("error")) {
+ return errResp(`Image generation failed: ${trimmed}`);
+ }
+ // Response is comma-separated base64 PNGs or URLs
+ const parts = trimmed
+ .split(",")
+ .map((s) => s.trim())
+ .filter(Boolean);
+ const images = parts.map((p) =>
+ p.startsWith("http") ? { url: p, type: "image" } : { b64_json: p, type: "image" }
+ );
+ return jsonResp({ object: "image.generation", data: images, status: "completed" });
+}
+
+async function handleTTS(prompt: string, voice?: string, lang?: string): Promise<Response> {
+ // Parse prompt for text and optional voice instructions
+ const text = prompt;
+ const selectedVoice = voice || "en-US-AvaNeural";
+ const selectedLang = lang || "en-US";
+
+ const res = await fetch(TTS_URL, {
+ method: "POST",
+ headers: {
+ "Content-Type": "application/json",
+ "User-Agent": USER_AGENT,
+ Origin: BASE_URL,
+ Referer: `${BASE_URL}/free-ai-text-to-speech/`,
+ },
+ body: JSON.stringify({
+ text: text.slice(0, 10000),
+ voice: selectedVoice,
+ lang: selectedLang,
+ pitch: "0",
+ speed: "1.0",
+ }),
+ });
+
+ if (!res.ok) {
+ return errResp(`TTS failed (${res.status})`);
+ }
+
+ const contentType = res.headers.get("content-type") || "";
+ if (
+ contentType.includes("audio") ||
+ contentType.includes("octet-stream") ||
+ contentType.includes("wav")
+ ) {
+ // Return audio directly
+ return new Response(res.body, {
+ headers: {
+ "Content-Type": contentType.includes("wav") ? "audio/wav" : "audio/mpeg",
+ "Content-Disposition": 'attachment; filename="speech.wav"',
+ },
+ });
+ }
+
+ // JSON response with base64 audio_data
+ const data = await res.text();
+ try {
+ const json = JSON.parse(data);
+ if (json.audio_data) {
+ return jsonResp({ object: "audio.speech", audio: json.audio_data, status: "completed" });
+ }
+ if (json.url) {
+ return jsonResp({ object: "audio.speech", url: json.url, status: "completed" });
+ }
+ } catch {
+ /* not JSON */
+ }
+ return errResp(`TTS unexpected response: ${data.slice(0, 200)}`);
+}
+
+async function handleEnhance(nonce: string, prompt: string): Promise<Response> {
+ const result = await postAjax(nonce, {
+ prompt,
+ actionType: "main-prompt-generation",
+ });
+ const trimmed = result.trim();
+ if (!trimmed || trimmed === "0") {
+ return errResp("Prompt enhancement failed");
+ }
+ return jsonResp({ object: "prompt.enhancement", enhanced: trimmed, status: "completed" });
+}
+
+// ─── Executor ───────────────────────────────────────────────────────────────
+
+export class VeoAIFreeWebExecutor extends BaseExecutor {
+ constructor() {
+ super("veoaifree-web", { id: "veoaifree-web", baseUrl: BASE_URL });
+ }
+
+ async execute(input: ExecuteInput): Promise<{
+ response: Response;
+ url: string;
+ headers: Record<string, string>;
+ transformedBody: unknown;
+ }> {
+ const body = input.body as Record<string, unknown> | undefined;
+ const model = input.model || (body?.model as string) || "veo-3.1";
+
+ // Extract prompt
+ const messages = (body?.messages as Array<Record<string, unknown>>) || [];
+ const userMsg = messages.filter((m) => m.role === "user").pop();
+ const systemMsg = messages.filter((m) => m.role === "system").pop();
+ const prompt = (userMsg?.content as string) || "";
+ const systemText = (systemMsg?.content as string) || "";
+
+ if (!prompt.trim()) {
+ return {
+ response: errResp("No prompt provided", 400),
+ url: AJAX_URL,
+ headers: {},
+ transformedBody: null,
+ };
+ }
+
+ // Detect intent
+ const intent = detectIntent(model, prompt);
+
+ // TTS doesn't need nonce
+ if (intent === "tts") {
+ const voiceMatch = systemText.match(/voice:\s*(\S+)/);
+ const langMatch = systemText.match(/lang:\s*(\S+)/);
+ const resp = await handleTTS(prompt, voiceMatch?.[1], langMatch?.[1]);
+ return { response: resp, url: TTS_URL, headers: {}, transformedBody: { intent, model } };
+ }
+
+ // Get nonce for AJAX endpoints
+ let nonce: string;
+ try {
+ nonce = await fetchNonce();
+ } catch (err) {
+ const msg = err instanceof Error ? err.message : "Failed to get nonce";
+ return { response: errResp(msg), url: BASE_URL, headers: {}, transformedBody: null };
+ }
+
+ // Extract aspect ratio from system prompt or default
+ const arMatch = systemText.match(/aspect[_-]?ratio:\s*(\S+)/i);
+ const aspectRatio = arMatch?.[1] || "VIDEO_ASPECT_RATIO_LANDSCAPE";
+
+ let resp: Response;
+ switch (intent) {
+ case "image":
+ resp = await handleImage(nonce, prompt, aspectRatio.replace("VIDEO_", "IMAGE_"));
+ break;
+ case "enhance":
+ resp = await handleEnhance(nonce, prompt);
+ break;
+ default:
+ resp = await handleVideo(nonce, prompt, aspectRatio);
+ }
+
+ return { response: resp, url: AJAX_URL, headers: {}, transformedBody: { intent, model } };
+ }
+}
diff --git a/src/shared/constants/providers.ts b/src/shared/constants/providers.ts
index 08f7fd648c..f1fc03e725 100644
--- a/src/shared/constants/providers.ts
+++ b/src/shared/constants/providers.ts
@@ -204,6 +204,18 @@ export const WEB_COOKIE_PROVIDERS = {
authHint:
"Paste your access_token from copilot.microsoft.com (or export a .har file from DevTools while logged in)",
},
+ "veoaifree-web": {
+ id: "veoaifree-web",
+ alias: "veo-free",
+ name: "Veo AI Free",
+ icon: "videocam",
+ color: "#8B5CF6",
+ textIcon: "VF",
+ website: "https://veoaifree.com",
+ hasFree: true,
+ freeNote: "Free video generation — VEO 3.1, Seedance. 6 requests/hour.",
+ authHint: "No auth required. Rate limited to 6 requests/hour per IP.",
+ },
};
// API Key Providers
@@ -1648,7 +1660,7 @@ export const ENTERPRISE_CLOUD_PROVIDER_IDS = new Set([
"modal",
]);
-export const VIDEO_PROVIDER_IDS = new Set(["runwayml"]);
+export const VIDEO_PROVIDER_IDS = new Set(["runwayml", "veoaifree-web"]);
export const EMBEDDING_RERANK_PROVIDER_IDS = new Set(["voyage-ai", "jina-ai"]);
diff --git a/tests/unit/veoaifree-web-executor.test.ts b/tests/unit/veoaifree-web-executor.test.ts
new file mode 100644
index 0000000000..38ff07dde6
--- /dev/null
+++ b/tests/unit/veoaifree-web-executor.test.ts
@@ -0,0 +1,75 @@
+import test from "node:test";
+import assert from "node:assert/strict";
+
+const { detectIntent } = await import("../../open-sse/executors/veoaifree-web.ts");
+
+// ─── detectIntent: model-based ──────────────────────────────────────────────
+
+test("detectIntent returns 'video' for veo models", () => {
+ assert.equal(detectIntent("veo-3.1"), "video");
+ assert.equal(detectIntent("veo-3.0"), "video");
+ assert.equal(detectIntent("seedance"), "video");
+ assert.equal(detectIntent("VEO-3.1"), "video");
+});
+
+test("detectIntent returns 'image' for image models", () => {
+ assert.equal(detectIntent("image-gen"), "image");
+ assert.equal(detectIntent("banana"), "image");
+ assert.equal(detectIntent("imagen-4"), "image");
+ assert.equal(detectIntent("nano-banana"), "image");
+});
+
+test("detectIntent returns 'tts' for audio models", () => {
+ assert.equal(detectIntent("tts"), "tts");
+ assert.equal(detectIntent("speech"), "tts");
+ assert.equal(detectIntent("audio-gen"), "tts");
+});
+
+test("detectIntent returns 'enhance' for prompt models", () => {
+ assert.equal(detectIntent("enhance"), "enhance");
+ assert.equal(detectIntent("prompt-helper"), "enhance");
+});
+
+// ─── detectIntent: prompt-based ─────────────────────────────────────────────
+
+test("detectIntent returns 'image' for image prompts", () => {
+ assert.equal(detectIntent(undefined, "generate image of a cat"), "image");
+ assert.equal(detectIntent(undefined, "create image of sunset"), "image");
+ assert.equal(detectIntent(undefined, "draw a horse"), "image");
+});
+
+test("detectIntent returns 'enhance' for enhance prompts", () => {
+ assert.equal(detectIntent(undefined, "enhance my prompt"), "enhance");
+ assert.equal(detectIntent(undefined, "improve prompt for video"), "enhance");
+});
+
+test("detectIntent defaults to 'video' for generic prompts", () => {
+ assert.equal(detectIntent(undefined, "a cat walking on the moon"), "video");
+ assert.equal(detectIntent(undefined, ""), "video");
+ assert.equal(detectIntent(undefined, undefined), "video");
+});
+
+// ─── detectIntent: case insensitive ─────────────────────────────────────────
+
+test("detectIntent is case-insensitive for model names", () => {
+ assert.equal(detectIntent("VEO-3.1"), "video");
+ assert.equal(detectIntent("TTS"), "tts");
+ assert.equal(detectIntent("Image-Gen"), "image");
+ assert.equal(detectIntent("ENHANCE"), "enhance");
+});
+
+// ─── detectIntent: model takes precedence over prompt ───────────────────────
+
+test("detectIntent model takes precedence over prompt", () => {
+ assert.equal(detectIntent("tts", "generate video of cat"), "tts");
+ assert.equal(detectIntent("veo-3.1", "create image"), "video");
+});
+
+// ─── Integration: executor class exists ─────────────────────────────────────
+
+test("VeoAIFreeWebExecutor class can be imported", async () => {
+ const { VeoAIFreeWebExecutor } = await import("../../open-sse/executors/veoaifree-web.ts");
+ const executor = new VeoAIFreeWebExecutor();
+ assert.ok(executor);
+ assert.equal(typeof executor.execute, "function");
+});