@fudrouter/fsrouter 0.6.64 → 0.6.66

package/open-sse/config/freeTierCatalog.js

@@ -0,0 +1,81 @@
+/**
+ * Free-tier monthly token budget catalog.
+ *
+ * Hand-seeded from the 2026-06-05 per-provider research snapshot.
+ */
+
+export const FREE_TIER_BUDGETS = {
+  mistral: 1_000_000_000,
+  "cloudflare-ai": 122_000_000,
+  gemini: 60_000_000,
+  doubao: 60_000_000,
+  cerebras: 30_000_000,
+  "api-airforce": 24_000_000,
+  "ollama-cloud": 20_000_000,
+  "github-models": 18_000_000,
+  groq: 15_000_000,
+  inclusionai: 15_000_000,
+  bluesminds: 7_200_000,
+  sambanova: 6_000_000,
+  "arcee-ai": 4_800_000,
+  llm7: 4_300_000,
+  bazaarlink: 3_600_000,
+  openrouter: 1_200_000,
+  cohere: 800_000,
+  huggingchat: 500_000,
+  morph: 400_000,
+  huggingface: 200_000,
+  kiro: 25_000,
+};
+
+/**
+ * Providers whose terms PROHIBIT routing through a self-hosted proxy or forbid
+ * non-personal use.
+ */
+export const FREE_TIER_TOS = {
+  opencode: "avoid",
+  "duckduckgo-web": "avoid",
+  agy: "avoid",
+  kiro: "avoid",
+  "amazon-q": "avoid",
+  "muse-spark-web": "avoid",
+  "t3-web": "avoid",
+  "qwen-web": "avoid",
+  modal: "avoid",
+  nlpcloud: "avoid",
+  blackbox: "avoid",
+  completions: "avoid",
+  fireworks: "avoid",
+  "featherless-ai": "avoid",
+  friendliai: "avoid",
+  ai21: "avoid",
+  iflytek: "avoid",
+  coze: "avoid",
+};
+
+function billions(n) {
+  return n >= 1e9 ? (n / 1e9).toFixed(2) + "B" : Math.round(n / 1e6) + "M";
+}
+
+/**
+ * Sum the documented free-tier budgets.
+ * @deprecated Superseded by the per-MODEL catalog. Use computeFreeModelTotals().
+ */
+export function computeFreeTierTotals(opts = {}) {
+  const byProvider = Object.entries(FREE_TIER_BUDGETS)
+    .map(([id, monthlyTokens]) => ({
+      id,
+      monthlyTokens,
+      tos: (FREE_TIER_TOS[id] ?? "caution"),
+    }))
+    .filter((p) => !(opts.excludeTosAvoid && p.tos === "avoid"))
+    .sort((a, b) => b.monthlyTokens - a.monthlyTokens);
+
+  const documentedMonthlyTokens = byProvider.reduce((s, p) => s + p.monthlyTokens, 0);
+  return {
+    documentedMonthlyTokens,
+    providerCount: byProvider.length,
+    byProvider,
+    headline: `over ${billions(documentedMonthlyTokens)} documented free tokens/month across ${byProvider.length}+ providers`,
+  };
+}

package/open-sse/config/googleTtsLanguages.js

@@ -0,0 +1,62 @@
+export const GOOGLE_TTS_LANGUAGES = [
+  { id: "af",    name: "Afrikaans",             type: "tts" },
+  { id: "ar",    name: "Arabic",                type: "tts" },
+  { id: "bg",    name: "Bulgarian",             type: "tts" },
+  { id: "bn",    name: "Bengali",               type: "tts" },
+  { id: "bs",    name: "Bosnian",               type: "tts" },
+  { id: "ca",    name: "Catalan",               type: "tts" },
+  { id: "cs",    name: "Czech",                 type: "tts" },
+  { id: "cy",    name: "Welsh",                 type: "tts" },
+  { id: "da",    name: "Danish",                type: "tts" },
+  { id: "de",    name: "German",                type: "tts" },
+  { id: "el",    name: "Greek",                 type: "tts" },
+  { id: "en",    name: "English",               type: "tts" },
+  { id: "eo",    name: "Esperanto",             type: "tts" },
+  { id: "es",    name: "Spanish",               type: "tts" },
+  { id: "et",    name: "Estonian",              type: "tts" },
+  { id: "fi",    name: "Finnish",               type: "tts" },
+  { id: "fr",    name: "French",                type: "tts" },
+  { id: "gu",    name: "Gujarati",              type: "tts" },
+  { id: "hi",    name: "Hindi",                 type: "tts" },
+  { id: "hr",    name: "Croatian",              type: "tts" },
+  { id: "hu",    name: "Hungarian",             type: "tts" },
+  { id: "hy",    name: "Armenian",              type: "tts" },
+  { id: "id",    name: "Indonesian",            type: "tts" },
+  { id: "is",    name: "Icelandic",             type: "tts" },
+  { id: "it",    name: "Italian",               type: "tts" },
+  { id: "ja",    name: "Japanese",              type: "tts" },
+  { id: "jw",    name: "Javanese",              type: "tts" },
+  { id: "km",    name: "Khmer",                 type: "tts" },
+  { id: "kn",    name: "Kannada",               type: "tts" },
+  { id: "ko",    name: "Korean",                type: "tts" },
+  { id: "la",    name: "Latin",                 type: "tts" },
+  { id: "lv",    name: "Latvian",               type: "tts" },
+  { id: "mk",    name: "Macedonian",            type: "tts" },
+  { id: "ml",    name: "Malayalam",             type: "tts" },
+  { id: "mr",    name: "Marathi",               type: "tts" },
+  { id: "my",    name: "Myanmar (Burmese)",      type: "tts" },
+  { id: "ne",    name: "Nepali",                type: "tts" },
+  { id: "nl",    name: "Dutch",                 type: "tts" },
+  { id: "no",    name: "Norwegian",             type: "tts" },
+  { id: "pl",    name: "Polish",                type: "tts" },
+  { id: "pt",    name: "Portuguese",            type: "tts" },
+  { id: "ro",    name: "Romanian",              type: "tts" },
+  { id: "ru",    name: "Russian",               type: "tts" },
+  { id: "si",    name: "Sinhala",               type: "tts" },
+  { id: "sk",    name: "Slovak",                type: "tts" },
+  { id: "sq",    name: "Albanian",              type: "tts" },
+  { id: "sr",    name: "Serbian",               type: "tts" },
+  { id: "su",    name: "Sundanese",             type: "tts" },
+  { id: "sv",    name: "Swedish",               type: "tts" },
+  { id: "sw",    name: "Swahili",               type: "tts" },
+  { id: "ta",    name: "Tamil",                 type: "tts" },
+  { id: "te",    name: "Telugu",                type: "tts" },
+  { id: "th",    name: "Thai",                  type: "tts" },
+  { id: "tl",    name: "Filipino",              type: "tts" },
+  { id: "tr",    name: "Turkish",               type: "tts" },
+  { id: "uk",    name: "Ukrainian",             type: "tts" },
+  { id: "ur",    name: "Urdu",                  type: "tts" },
+  { id: "vi",    name: "Vietnamese",            type: "tts" },
+  { id: "zh-CN", name: "Chinese (Simplified)",  type: "tts" },
+  { id: "zh-TW", name: "Chinese (Traditional)", type: "tts" },
+];

package/open-sse/config/imageParamDefs.js

@@ -0,0 +1,360 @@
+/**
+ * Image Generation — Canonical Parameter Schema Library
+ *
+ * This file is the single source of truth for all parameter definitions
+ * used by image (and video) generation models across every provider.
+ *
+ * Each entry in IMAGE_PARAMS fully describes one request parameter:
+ *   name        — param key as sent in the API body
+ *   type        — "string" | "integer" | "number" | "boolean"
+ *   required    — whether the API will reject requests missing this param
+ *   default     — value used when param is omitted (null = provider decides)
+ *   validValues — exhaustive list of accepted string values (null = freeform)
+ *   min / max   — numeric bounds (only for type "integer" / "number")
+ *   description — human-readable explanation shown in UI and docs
+ *
+ * Usage:
+ *   import { pd, p } from "./imageParamDefs.js";
+ *
+ *   // Build paramDefs from a list of names (with optional per-name overrides)
+ *   pd(["size", "n", "quality"])
+ *   pd(["size"], { size: { validValues: ["1:1", "16:9"] } })
+ *
+ *   // Get a single param def (with optional overrides)
+ *   p("size")
+ *   p("size", { validValues: ["1024x1024", "1792x1024"] })
+ */
+
+// ─── Standard OpenAI size strings ────────────────────────────────────────────
+const STD_SIZES   = ["1024x1024", "1792x1024", "1024x1792", "1024x1536", "1536x1024"];
+const DALLE3_SIZES = ["1024x1024", "1792x1024", "1024x1792"];
+const DALLE2_SIZES = ["256x256", "512x512", "1024x1024"];
+const ASPECT_RATIOS = ["1:1", "16:9", "9:16", "4:3", "3:4", "3:2", "2:3"];
+
+// ─── Canonical param definitions ─────────────────────────────────────────────
+export const IMAGE_PARAMS = {
+
+  // ── Dimensions ──────────────────────────────────────────────────────────────
+
+  size: {
+    name: "size",
+    type: "string",
+    required: false,
+    default: "1024x1024",
+    validValues: STD_SIZES,
+    description: "Output image dimensions as 'WIDTHxHEIGHT'. Standard: 1024×1024 (square), 1792×1024 (landscape), 1024×1792 (portrait).",
+  },
+
+  aspect_ratio: {
+    name: "aspect_ratio",
+    type: "string",
+    required: false,
+    default: "1:1",
+    validValues: ASPECT_RATIOS,
+    description: "Output aspect ratio. Used by providers that accept ratio strings instead of pixel dimensions.",
+  },
+
+  // ── Quantity ─────────────────────────────────────────────────────────────────
+
+  n: {
+    name: "n",
+    type: "integer",
+    required: false,
+    default: 1,
+    min: 1,
+    max: 4,
+    description: "Number of images to generate in a single request (1–4).",
+  },
+
+  // ── Quality & style ──────────────────────────────────────────────────────────
+
+  quality: {
+    name: "quality",
+    type: "string",
+    required: false,
+    default: "standard",
+    validValues: ["standard", "hd", "low", "medium", "high", "auto"],
+    description: "Image rendering quality. Higher quality produces more detailed images but may take longer. Provider-specific values apply.",
+  },
+
+  style: {
+    name: "style",
+    type: "string",
+    required: false,
+    default: null,
+    validValues: ["vivid", "natural", "anime", "realistic", "digital-art"],
+    description: "Artistic style preset. 'vivid' produces hyper-real images; 'natural' is more subdued. Provider-specific presets may apply.",
+  },
+
+  output_format: {
+    name: "output_format",
+    type: "string",
+    required: false,
+    default: "png",
+    validValues: ["png", "jpeg", "webp"],
+    description: "Output image file format. 'png' supports transparency; 'jpeg'/'webp' are smaller for photos.",
+  },
+
+  // ── Response format ──────────────────────────────────────────────────────────
+
+  response_format: {
+    name: "response_format",
+    type: "string",
+    required: false,
+    default: "url",
+    validValues: ["url", "b64_json"],
+    description: "How the image is returned: 'url' is a temporary public link; 'b64_json' is the raw image as a base64-encoded JSON string.",
+  },
+
+  // ── Reference inputs ─────────────────────────────────────────────────────────
+
+  image_url: {
+    name: "image_url",
+    type: "string",
+    required: false,
+    default: null,
+    description: "Reference image for img2img / conditioning. Accepts an HTTP URL or a base64 Data URI (data:image/png;base64,…). Used as the start frame or style reference.",
+  },
+
+  end_image_url: {
+    name: "end_image_url",
+    type: "string",
+    required: false,
+    default: null,
+    description: "End/last-frame image URL for first-last-frame video models (e.g. Kling, Veo). The model interpolates video between image_url (start) and end_image_url (end).",
+  },
+
+  video_url: {
+    name: "video_url",
+    type: "string",
+    required: false,
+    default: null,
+    description: "Reference video URL for video-to-video conditioning (e.g. motion transfer, style reference on video input).",
+  },
+
+  image: {
+    name: "image",
+    type: "string",
+    required: false,
+    default: null,
+    description: "Input image for editing or conditioning. Accepts an HTTP URL or base64 Data URI. Identical in function to image_url — some providers use this key instead.",
+  },
+
+  mask_image: {
+    name: "mask_image",
+    type: "string",
+    required: false,
+    default: null,
+    description: "Mask image for inpainting. White areas = regions to regenerate; black areas = regions to preserve. Accepts HTTP URL or base64 Data URI.",
+  },
+
+  negative_prompt: {
+    name: "negative_prompt",
+    type: "string",
+    required: false,
+    default: null,
+    description: "Text describing elements to exclude from the generated image (e.g. 'blurry, low quality, watermark').",
+  },
+
+  // ── Advanced diffusion controls ──────────────────────────────────────────────
+
+  seed: {
+    name: "seed",
+    type: "integer",
+    required: false,
+    default: null,
+    description: "Fixed random seed for reproducible results. Same seed + same prompt → same image (when other settings are equal).",
+  },
+
+  guidance: {
+    name: "guidance",
+    type: "number",
+    required: false,
+    default: null,
+    min: 0,
+    max: 20,
+    description: "Classifier-free guidance scale. Higher values make the output follow the prompt more closely but may reduce visual quality. Typical range: 5–15.",
+  },
+
+  num_steps: {
+    name: "num_steps",
+    type: "integer",
+    required: false,
+    default: null,
+    min: 1,
+    max: 50,
+    description: "Number of diffusion sampling steps. More steps = higher quality, slower generation. Typical range: 20–50.",
+  },
+
+  strength: {
+    name: "strength",
+    type: "number",
+    required: false,
+    default: null,
+    min: 0,
+    max: 1,
+    description: "Img2img denoising strength. 0 = identical to input; 1 = completely new image. Typical range: 0.5–0.9.",
+  },
+
+  // ── Codex / GPT-Image specific ───────────────────────────────────────────────
+
+  background: {
+    name: "background",
+    type: "string",
+    required: false,
+    default: "auto",
+    validValues: ["auto", "transparent", "opaque"],
+    description: "Background mode (Codex/GPT Image only). 'transparent' requires output_format=png.",
+  },
+
+  image_detail: {
+    name: "image_detail",
+    type: "string",
+    required: false,
+    default: "high",
+    validValues: ["low", "high", "auto"],
+    description: "Detail level for input reference images (Codex/GPT Image only). 'high' uses more tokens but preserves fine detail.",
+  },
+
+  // ── Video-specific params ────────────────────────────────────────────────────
+
+  duration: {
+    name: "duration",
+    type: "integer",
+    required: false,
+    default: 5,
+    min: 1,
+    max: 30,
+    description: "Video duration in seconds. Supported range varies by model (e.g. 5–15s for Kling, fixed 8s for some Veo models).",
+  },
+
+  resolution: {
+    name: "resolution",
+    type: "string",
+    required: false,
+    default: null,
+    validValues: ["480p", "720p", "1080p", "4k"],
+    description: "Video output resolution. Higher resolutions require more generation time and cost.",
+  },
+
+  motion_strength: {
+    name: "motion_strength",
+    type: "number",
+    required: false,
+    default: null,
+    min: 0,
+    max: 1,
+    description: "Camera/subject motion intensity for video generation. 0 = static / minimal movement; 1 = high dynamic motion.",
+  },
+
+  video_url: {
+    name: "video_url",
+    type: "string",
+    required: false,
+    default: null,
+    description: "Reference video URL or base64 for video-to-video conditioning (start/end frame or motion reference).",
+  },
+};
+
+// ─── Helpers ──────────────────────────────────────────────────────────────────
+
+/**
+ * Return the canonical param definition for a single param name,
+ * with optional field-level overrides.
+ *
+ * @param {string} name - key in IMAGE_PARAMS
+ * @param {object} [overrides] - partial fields to override on the base definition
+ * @returns {object|null}
+ */
+export function p(name, overrides = {}) {
+  const base = IMAGE_PARAMS[name];
+  if (!base) return null;
+  if (!overrides || Object.keys(overrides).length === 0) return base;
+  return { ...base, ...overrides };
+}
+
+/**
+ * Build a paramDefs array from a list of param names.
+ * Unknown names are silently skipped.
+ *
+ * @param {string[]} names - list of param keys
+ * @param {object}  [overrides] - map of { paramName: { field overrides } }
+ * @returns {object[]}
+ */
+export function pd(names, overrides = {}) {
+  return names
+    .map(name => {
+      const base = IMAGE_PARAMS[name];
+      if (!base) return null;
+      const ov = overrides[name];
+      return ov ? { ...base, ...ov } : base;
+    })
+    .filter(Boolean);
+}
+
+// ─── Shared override groups (reused across providers) ─────────────────────────
+
+/**
+ * Weavy image models — frontend sends standard ratio strings (e.g. "9:16").
+ * The backend (weavy_service.py) converts to FAL-style image_size internally.
+ * Apply as: pd(["size", ...], WEAVY_SIZE_OV)
+ */
+export const WEAVY_SIZE_OV = {
+  size: {
+    default: "1:1",
+    validValues: ["1:1", "9:16", "16:9", "3:4", "4:3"],
+    description: "Output aspect ratio. Weavy converts this to the correct format per model (FAL image_size, Replicate aspect_ratio, or GPT Image pixel size).",
+  },
+};
+
+/**
+ * Standard aspect-ratio-only size param for providers that don't use WxH.
+ */
+export const RATIO_SIZE_OV = {
+  size: {
+    validValues: ASPECT_RATIOS,
+    description: "Output aspect ratio (e.g. '1:1', '16:9', '9:16').",
+  },
+};
+
+/**
+ * DALL-E 3 size restrictions.
+ */
+export const DALLE3_SIZE_OV = {
+  size: { validValues: DALLE3_SIZES },
+};
+
+/**
+ * DALL-E 2 size restrictions.
+ */
+export const DALLE2_SIZE_OV = {
+  size: { validValues: DALLE2_SIZES },
+};
+
+/**
+ * GPT Image 1 quality options.
+ */
+export const GPT_IMAGE_QUALITY_OV = {
+  quality: { validValues: ["low", "medium", "high", "auto"], default: "auto" },
+};
+
+/**
+ * Codex quality options.
+ */
+export const CODEX_QUALITY_OV = {
+  quality: { validValues: ["low", "medium", "high", "auto"], default: "auto" },
+};
+
+/**
+ * DALL-E 3 quality options.
+ */
+export const DALLE3_QUALITY_OV = {
+  quality: { validValues: ["standard", "hd"], default: "standard" },
+};
+
+/**
+ * n limited to 1 (provider does not support batch generation).
+ */
+export const N1_OV = {
+  n: { max: 1, description: "This model always generates exactly 1 image per request." },
+};