Refactor error handling to config-driven approach with centralized error rules

Made-with: Cursor
This commit is contained in:
decolua 2026-04-15 10:45:46 +07:00
parent b1288c5064
commit b669b6ffc1
20 changed files with 1056 additions and 991 deletions

View file

@ -0,0 +1,82 @@
// OpenAI-compatible error types mapping (client-facing)
export const ERROR_TYPES = {
400: { type: "invalid_request_error", code: "bad_request" },
401: { type: "authentication_error", code: "invalid_api_key" },
402: { type: "billing_error", code: "payment_required" },
403: { type: "permission_error", code: "insufficient_quota" },
404: { type: "invalid_request_error", code: "model_not_found" },
406: { type: "invalid_request_error", code: "model_not_supported" },
429: { type: "rate_limit_error", code: "rate_limit_exceeded" },
500: { type: "server_error", code: "internal_server_error" },
502: { type: "server_error", code: "bad_gateway" },
503: { type: "server_error", code: "service_unavailable" },
504: { type: "server_error", code: "gateway_timeout" }
};
// Default error messages per status code (client-facing)
export const DEFAULT_ERROR_MESSAGES = {
400: "Bad request",
401: "Invalid API key provided",
402: "Payment required",
403: "You exceeded your current quota",
404: "Model not found",
406: "Model not supported",
429: "Rate limit exceeded",
500: "Internal server error",
502: "Bad gateway - upstream provider error",
503: "Service temporarily unavailable",
504: "Gateway timeout"
};
// Exponential backoff config for rate limits
export const BACKOFF_CONFIG = {
base: 1000,
max: 3 * 60 * 1000,
maxLevel: 15
};
// Default cooldown for transient/unknown errors
export const TRANSIENT_COOLDOWN_MS = 30 * 1000;
// Cooldown durations (ms)
const COOLDOWN = {
long: 2 * 60 * 1000,
short: 5 * 1000,
};
/**
* Unified error classification rules.
* Checked top-to-bottom: text rules first (by order), then status rules.
* Each rule: { text?, status?, cooldownMs?, backoff? }
* - text: substring match (case-insensitive) on error message
* - status: HTTP status code match
* - cooldownMs: fixed cooldown duration
* - backoff: true = use exponential backoff (rate limit)
*/
export const ERROR_RULES = [
// --- Text-based rules (checked first, order = priority) ---
{ text: "no credentials", cooldownMs: COOLDOWN.long },
{ text: "request not allowed", cooldownMs: COOLDOWN.short },
{ text: "improperly formed request", cooldownMs: COOLDOWN.long },
{ text: "rate limit", backoff: true },
{ text: "too many requests", backoff: true },
{ text: "quota exceeded", backoff: true },
{ text: "capacity", backoff: true },
{ text: "overloaded", backoff: true },
// --- Status-based rules (fallback when text doesn't match) ---
{ status: 401, cooldownMs: COOLDOWN.long },
{ status: 402, cooldownMs: COOLDOWN.long },
{ status: 403, cooldownMs: COOLDOWN.long },
{ status: 404, cooldownMs: COOLDOWN.long },
{ status: 429, backoff: true },
];
// Backward compat: COOLDOWN_MS object (used by index.js re-export)
export const COOLDOWN_MS = {
unauthorized: COOLDOWN.long,
paymentRequired: COOLDOWN.long,
notFound: COOLDOWN.long,
transient: TRANSIENT_COOLDOWN_MS,
requestNotAllowed: COOLDOWN.short,
};

View file

@ -1,5 +1,5 @@
import { PROVIDERS } from "./providers.js";
import { GOOGLE_TTS_LANGUAGES } from "./googleTtsLanguages.js";
import { buildTtsProviderModels } from "./ttsModels.js";
// Provider models - Single source of truth
// Key = alias (cc, cx, gc, qw, if, ag, gh for OAuth; id for API Key)
@ -144,10 +144,10 @@ export const PROVIDER_MODELS = {
{ id: "deepseek/deepseek-reasoner", name: "DeepSeek Reasoner" },
],
oc: [ // OpenCode
{ id: "nemotron-3-super-free", name: "Nemotron 3 Super" },
// { id: "nemotron-3-super-free", name: "Nemotron 3 Super" },
// { id: "qwen3.6-plus-free", name: "Qwen 3.6 Plus" },
// { id: "big-pickle", name: "Big Pickle", targetFormat: "claude" },
{ id: "minimax-m2.5-free", name: "MiniMax M2.5", targetFormat: "claude" },
// { id: "minimax-m2.5-free", name: "MiniMax M2.5", targetFormat: "claude" },
// { id: "trinity-large-preview-free", name: "Trinity Large Preview" },
],
@ -230,6 +230,10 @@ export const PROVIDER_MODELS = {
{ id: "perplexity/pplx-embed-v1-4b", name: "Perplexity Embed V1 4B", type: "embedding" },
{ id: "perplexity/pplx-embed-v1-0.6b", name: "Perplexity Embed V1 0.6B", type: "embedding" },
{ id: "nvidia/llama-nemotron-embed-vl-1b-v2:free", name: "NVIDIA Nemotron Embed VL 1B V2 (Free)", type: "embedding" },
// TTS models
{ id: "openai/gpt-4o-mini-tts", name: "GPT-4o Mini TTS", type: "tts" },
{ id: "openai/tts-1-hd", name: "TTS-1 HD", type: "tts" },
{ id: "openai/tts-1", name: "TTS-1", type: "tts" },
],
glm: [
{ id: "glm-5.1", name: "GLM 5.1" },
@ -377,54 +381,8 @@ export const PROVIDER_MODELS = {
{ id: "zai-org/glm-5-maas", name: "GLM-5 (Vertex)" },
],
// Free/noAuth TTS providers
"local-device": [
{ id: "default", name: "System Default Voice", type: "tts" },
],
"google-tts": GOOGLE_TTS_LANGUAGES,
// OpenAI TTS voices (hardcoded — no public API to list them)
// Used by ttsCore.js when provider = openai
"openai-tts-voices": [
{ id: "alloy", name: "Alloy", type: "tts" },
{ id: "ash", name: "Ash", type: "tts" },
{ id: "ballad", name: "Ballad", type: "tts" },
{ id: "cedar", name: "Cedar", type: "tts" },
{ id: "coral", name: "Coral", type: "tts" },
{ id: "echo", name: "Echo", type: "tts" },
{ id: "fable", name: "Fable", type: "tts" },
{ id: "marin", name: "Marin", type: "tts" },
{ id: "nova", name: "Nova", type: "tts" },
{ id: "onyx", name: "Onyx", type: "tts" },
{ id: "sage", name: "Sage", type: "tts" },
{ id: "shimmer", name: "Shimmer", type: "tts" },
{ id: "verse", name: "Verse", type: "tts" },
],
// OpenAI TTS models
"openai-tts-models": [
{ id: "gpt-4o-mini-tts", name: "GPT-4o Mini TTS", type: "tts" },
{ id: "tts-1-hd", name: "TTS-1 HD", type: "tts" },
{ id: "tts-1", name: "TTS-1", type: "tts" },
],
// ElevenLabs TTS models
"elevenlabs-tts-models": [
{ id: "eleven_flash_v2_5", name: "Flash v2.5 (Fastest)", type: "tts" },
{ id: "eleven_turbo_v2_5", name: "Turbo v2.5 (Fast)", type: "tts" },
{ id: "eleven_multilingual_v2", name: "Multilingual v2 (Quality)", type: "tts" },
{ id: "eleven_monolingual_v1", name: "Monolingual v1 (English)", type: "tts" },
],
"edge-tts": [
{ id: "en-US-AriaNeural", name: "Aria (en-US)", type: "tts" },
{ id: "en-US-GuyNeural", name: "Guy (en-US)", type: "tts" },
{ id: "en-GB-SoniaNeural", name: "Sonia (en-GB)", type: "tts" },
{ id: "vi-VN-HoaiMyNeural", name: "Hoai My (vi-VN)", type: "tts" },
{ id: "vi-VN-NamMinhNeural", name: "Nam Minh (vi-VN)", type: "tts" },
{ id: "zh-CN-XiaoxiaoNeural", name: "Xiaoxiao (zh-CN)", type: "tts" },
{ id: "zh-CN-YunxiNeural", name: "Yunxi (zh-CN)", type: "tts" },
{ id: "fr-FR-DeniseNeural", name: "Denise (fr-FR)", type: "tts" },
{ id: "de-DE-KatjaNeural", name: "Katja (de-DE)", type: "tts" },
{ id: "ja-JP-NanamiNeural", name: "Nanami (ja-JP)", type: "tts" },
{ id: "ko-KR-SunHiNeural", name: "SunHi (ko-KR)", type: "tts" },
],
// TTS entries are loaded from ttsModels.js via buildTtsProviderModels()
...buildTtsProviderModels(),
};
// Helper functions

View file

@ -14,33 +14,8 @@ export const HTTP_STATUS = {
GATEWAY_TIMEOUT: 504
};
// OpenAI-compatible error types mapping
export const ERROR_TYPES = {
[HTTP_STATUS.BAD_REQUEST]: { type: "invalid_request_error", code: "bad_request" },
[HTTP_STATUS.UNAUTHORIZED]: { type: "authentication_error", code: "invalid_api_key" },
[HTTP_STATUS.FORBIDDEN]: { type: "permission_error", code: "insufficient_quota" },
[HTTP_STATUS.NOT_FOUND]: { type: "invalid_request_error", code: "model_not_found" },
[HTTP_STATUS.NOT_ACCEPTABLE]: { type: "invalid_request_error", code: "model_not_supported" },
[HTTP_STATUS.RATE_LIMITED]: { type: "rate_limit_error", code: "rate_limit_exceeded" },
[HTTP_STATUS.SERVER_ERROR]: { type: "server_error", code: "internal_server_error" },
[HTTP_STATUS.BAD_GATEWAY]: { type: "server_error", code: "bad_gateway" },
[HTTP_STATUS.SERVICE_UNAVAILABLE]: { type: "server_error", code: "service_unavailable" },
[HTTP_STATUS.GATEWAY_TIMEOUT]: { type: "server_error", code: "gateway_timeout" }
};
// Default error messages per status code
export const DEFAULT_ERROR_MESSAGES = {
[HTTP_STATUS.BAD_REQUEST]: "Bad request",
[HTTP_STATUS.UNAUTHORIZED]: "Invalid API key provided",
[HTTP_STATUS.FORBIDDEN]: "You exceeded your current quota",
[HTTP_STATUS.NOT_FOUND]: "Model not found",
[HTTP_STATUS.NOT_ACCEPTABLE]: "Model not supported",
[HTTP_STATUS.RATE_LIMITED]: "Rate limit exceeded",
[HTTP_STATUS.SERVER_ERROR]: "Internal server error",
[HTTP_STATUS.BAD_GATEWAY]: "Bad gateway - upstream provider error",
[HTTP_STATUS.SERVICE_UNAVAILABLE]: "Service temporarily unavailable",
[HTTP_STATUS.GATEWAY_TIMEOUT]: "Gateway timeout"
};
// Re-export error config (backward compat)
export { ERROR_TYPES, DEFAULT_ERROR_MESSAGES, BACKOFF_CONFIG, COOLDOWN_MS } from "./errorConfig.js";
// Cache TTLs (seconds)
export const CACHE_TTL = {
@ -73,26 +48,6 @@ export const DEFAULT_RETRY_CONFIG = {
502: 1 // Bad gateway - retry 1 time (transient)
};
// Exponential backoff config for rate limits
export const BACKOFF_CONFIG = {
base: 1000,
max: 2 * 60 * 1000,
maxLevel: 15
};
// Error-based cooldown times
export const COOLDOWN_MS = {
unauthorized: 2 * 60 * 1000,
paymentRequired: 2 * 60 * 1000,
notFound: 2 * 60 * 1000,
transient: 30 * 1000,
requestNotAllowed: 5 * 1000,
// Legacy aliases
rateLimit: 2 * 60 * 1000,
serviceUnavailable: 2 * 1000,
authExpired: 2 * 60 * 1000
};
// Requests containing these texts will bypass provider
export const SKIP_PATTERNS = [
"Please write a 5-10 word title for the following conversation:"

View file

@ -0,0 +1,109 @@
import { GOOGLE_TTS_LANGUAGES } from "./googleTtsLanguages.js";
// ── Voice definitions (DRY — reused across providers) ──────────────────────
const VOICES = {
alloy: { id: "alloy", name: "Alloy" },
ash: { id: "ash", name: "Ash" },
ballad: { id: "ballad", name: "Ballad" },
cedar: { id: "cedar", name: "Cedar" },
coral: { id: "coral", name: "Coral" },
echo: { id: "echo", name: "Echo" },
fable: { id: "fable", name: "Fable" },
marin: { id: "marin", name: "Marin" },
nova: { id: "nova", name: "Nova" },
onyx: { id: "onyx", name: "Onyx" },
sage: { id: "sage", name: "Sage" },
shimmer: { id: "shimmer", name: "Shimmer" },
verse: { id: "verse", name: "Verse" },
};
const v = (...keys) => keys.map((k) => ({ ...VOICES[k], type: "tts" }));
// 9 voices for tts-1 / tts-1-hd
const VOICES_STANDARD = v("alloy", "ash", "coral", "echo", "fable", "nova", "onyx", "sage", "shimmer");
// 13 voices for gpt-4o-mini-tts
const VOICES_FULL = v("alloy", "ash", "ballad", "cedar", "coral", "echo", "fable", "marin", "nova", "onyx", "sage", "shimmer", "verse");
// ── TTS Config (config-driven, single source of truth) ─────────────────────
export const TTS_MODELS_CONFIG = {
openai: {
models: [
{ id: "gpt-4o-mini-tts", name: "GPT-4o Mini TTS", type: "tts" },
{ id: "tts-1-hd", name: "TTS-1 HD", type: "tts" },
{ id: "tts-1", name: "TTS-1", type: "tts" },
],
voices: {
"gpt-4o-mini-tts": VOICES_FULL,
"tts-1": VOICES_STANDARD,
"tts-1-hd": VOICES_STANDARD,
},
// Flat voice list (all unique voices) for backward compat
allVoices: VOICES_FULL,
},
openrouter: {
models: [
{ id: "openai/gpt-4o-mini-tts", name: "GPT-4o Mini TTS", type: "tts" },
{ id: "openai/tts-1-hd", name: "TTS-1 HD", type: "tts" },
{ id: "openai/tts-1", name: "TTS-1", type: "tts" },
],
voices: {
"openai/gpt-4o-mini-tts": VOICES_FULL,
"openai/tts-1": VOICES_STANDARD,
"openai/tts-1-hd": VOICES_STANDARD,
},
allVoices: VOICES_FULL,
},
elevenlabs: {
models: [
{ id: "eleven_flash_v2_5", name: "Flash v2.5 (Fastest)", type: "tts" },
{ id: "eleven_turbo_v2_5", name: "Turbo v2.5 (Fast)", type: "tts" },
{ id: "eleven_multilingual_v2", name: "Multilingual v2 (Quality)", type: "tts" },
{ id: "eleven_monolingual_v1", name: "Monolingual v1 (English)", type: "tts" },
],
// voices come from API, not hardcoded
},
"edge-tts": {
defaults: [
{ id: "en-US-AriaNeural", name: "Aria (en-US)", type: "tts" },
{ id: "en-US-GuyNeural", name: "Guy (en-US)", type: "tts" },
{ id: "en-GB-SoniaNeural", name: "Sonia (en-GB)", type: "tts" },
{ id: "vi-VN-HoaiMyNeural", name: "Hoai My (vi-VN)", type: "tts" },
{ id: "vi-VN-NamMinhNeural", name: "Nam Minh (vi-VN)", type: "tts" },
{ id: "zh-CN-XiaoxiaoNeural", name: "Xiaoxiao (zh-CN)", type: "tts" },
{ id: "zh-CN-YunxiNeural", name: "Yunxi (zh-CN)", type: "tts" },
{ id: "fr-FR-DeniseNeural", name: "Denise (fr-FR)", type: "tts" },
{ id: "de-DE-KatjaNeural", name: "Katja (de-DE)", type: "tts" },
{ id: "ja-JP-NanamiNeural", name: "Nanami (ja-JP)", type: "tts" },
{ id: "ko-KR-SunHiNeural", name: "SunHi (ko-KR)", type: "tts" },
],
},
"local-device": {
defaults: [
{ id: "default", name: "System Default Voice", type: "tts" },
],
},
"google-tts": {
defaults: GOOGLE_TTS_LANGUAGES,
},
};
// ── Helper: get voices for a specific model ────────────────────────────────
export function getTtsVoicesForModel(provider, modelId) {
const cfg = TTS_MODELS_CONFIG[provider];
if (!cfg?.voices) return null;
return cfg.voices[modelId] || cfg.allVoices || null;
}
// ── Build flat entries for PROVIDER_MODELS backward compat ─────────────────
export function buildTtsProviderModels() {
const entries = {};
for (const [provider, cfg] of Object.entries(TTS_MODELS_CONFIG)) {
if (cfg.models) entries[`${provider}-tts-models`] = cfg.models;
if (cfg.allVoices) entries[`${provider}-tts-voices`] = cfg.allVoices;
if (cfg.defaults) entries[provider] = cfg.defaults;
}
// Keep openai-tts-voices key pointing to full voice list for backward compat
entries["openai-tts-voices"] = TTS_MODELS_CONFIG.openai.allVoices;
entries["openrouter-tts-voices"] = TTS_MODELS_CONFIG.openrouter.allVoices;
return entries;
}

View file

@ -2,7 +2,7 @@ import { BaseExecutor } from "./base.js";
import { PROVIDERS } from "../config/providers.js";
// Models that use /zen/v1/messages (claude format)
const MESSAGES_MODELS = new Set(["big-pickle", "minimax-m2.5-free"]);
const MESSAGES_MODELS = new Set(["big-pickle"]);
export class OpenCodeExecutor extends BaseExecutor {
constructor() {

View file

@ -197,7 +197,7 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
// Provider returned error
if (!providerResponse.ok) {
trackPendingRequest(model, provider, connectionId, false, true);
const { statusCode, message, retryAfterMs } = await parseUpstreamError(providerResponse, provider);
const { statusCode, message } = await parseUpstreamError(providerResponse);
appendRequestLog({ model, provider, connectionId, status: `FAILED ${statusCode}` }).catch(() => {});
saveRequestDetail(buildRequestDetail({
provider, model, connectionId,
@ -211,11 +211,8 @@ export async function handleChatCore({ body, modelInfo, credentials, log, onCred
const errMsg = formatProviderError(new Error(message), provider, model, statusCode);
console.log(`${COLORS.red}[ERROR] ${errMsg}${COLORS.reset}`);
if (retryAfterMs && provider === "antigravity") {
log?.debug?.("RETRY", `Antigravity quota reset in ${Math.ceil(retryAfterMs / 1000)}s`);
}
reqLogger.logError(new Error(message), finalBody || translatedBody);
return createErrorResult(statusCode, errMsg, retryAfterMs);
return createErrorResult(statusCode, errMsg);
}
const sharedCtx = { provider, model, body, stream, translatedBody, finalBody, requestStartTime, connectionId, apiKey, clientRawRequest, onRequestSuccess };

View file

@ -270,7 +270,7 @@ export async function handleEmbeddingsCore({
}
if (!providerResponse.ok) {
const { statusCode, message } = await parseUpstreamError(providerResponse, provider);
const { statusCode, message } = await parseUpstreamError(providerResponse);
const errMsg = formatProviderError(new Error(message), provider, model, statusCode);
log?.debug?.("EMBEDDINGS", `Provider error: ${errMsg}`);
return createErrorResult(statusCode, errMsg);

View file

@ -339,6 +339,84 @@ export const VOICE_FETCHERS = {
// openai: uses hardcoded voices from providerModels.js
};
// ── OpenRouter TTS (via chat completions + audio modality) ───────────────────
async function handleOpenRouterTts({ model, input, credentials, responseFormat = "mp3" }) {
if (!credentials?.apiKey) {
return createErrorResult(HTTP_STATUS.UNAUTHORIZED, "No OpenRouter API key configured");
}
// model format: "tts-model/voice" e.g. "openai/gpt-4o-mini-tts/alloy"
let ttsModel = "openai/gpt-4o-mini-tts";
let voice = "alloy";
if (model && model.includes("/")) {
const lastSlash = model.lastIndexOf("/");
const maybVoice = model.slice(lastSlash + 1);
const maybeModel = model.slice(0, lastSlash);
// voice names are simple lowercase words, model names contain "/"
if (maybeModel.includes("/")) {
ttsModel = maybeModel;
voice = maybVoice;
} else {
voice = model;
}
} else if (model) {
voice = model;
}
const res = await fetch("https://openrouter.ai/api/v1/chat/completions", {
method: "POST",
headers: {
"Content-Type": "application/json",
"Authorization": `Bearer ${credentials.apiKey}`,
"HTTP-Referer": "https://endpoint-proxy.local",
"X-Title": "Endpoint Proxy",
},
body: JSON.stringify({
model: ttsModel,
modalities: ["text", "audio"],
audio: { voice, format: "wav" },
stream: true,
messages: [{ role: "user", content: input }],
}),
});
if (!res.ok) {
const err = await res.json().catch(() => ({}));
return createErrorResult(res.status, err?.error?.message || `OpenRouter TTS failed: ${res.status}`);
}
// Parse SSE stream, accumulate base64 audio chunks
const chunks = [];
const reader = res.body.getReader();
const decoder = new TextDecoder();
let buffer = "";
while (true) {
const { done, value } = await reader.read();
if (done) break;
buffer += decoder.decode(value, { stream: true });
const lines = buffer.split("\n");
buffer = lines.pop();
for (const line of lines) {
if (!line.startsWith("data: ") || line === "data: [DONE]") continue;
try {
const json = JSON.parse(line.slice(6));
const audioData = json.choices?.[0]?.delta?.audio?.data;
if (audioData) chunks.push(audioData);
} catch {}
}
}
if (chunks.length === 0) {
return createErrorResult(HTTP_STATUS.BAD_GATEWAY, "OpenRouter TTS returned no audio data");
}
const base64Audio = chunks.join("");
return createTtsResponse(base64Audio, "wav", responseFormat);
}
// ── OpenAI TTS ───────────────────────────────────────────────────────────────
async function handleOpenAiTts({ model, input, credentials, responseFormat = "mp3" }) {
if (!credentials?.apiKey) {
@ -422,6 +500,12 @@ const TTS_PROVIDERS = {
},
requiresCredentials: true,
},
"openrouter": {
synthesize: async (text, model, credentials, responseFormat) => {
return await handleOpenRouterTts({ model, input: text, credentials, responseFormat });
},
requiresCredentials: true,
},
};
// ── Core handler ───────────────────────────────────────────────

View file

@ -1,4 +1,4 @@
import { COOLDOWN_MS, BACKOFF_CONFIG, HTTP_STATUS } from "../config/runtimeConfig.js";
import { ERROR_RULES, BACKOFF_CONFIG, TRANSIENT_COOLDOWN_MS } from "../config/errorConfig.js";
/**
* Calculate exponential backoff cooldown for rate limits (429)
@ -13,82 +13,39 @@ export function getQuotaCooldown(backoffLevel = 0) {
/**
* Check if error should trigger account fallback (switch to next account)
* Config-driven: matches ERROR_RULES top-to-bottom (text rules first, then status)
* @param {number} status - HTTP status code
* @param {string} errorText - Error message text
* @param {number} backoffLevel - Current backoff level for exponential backoff
* @returns {{ shouldFallback: boolean, cooldownMs: number, newBackoffLevel?: number }}
*/
export function checkFallbackError(status, errorText, backoffLevel = 0) {
// Check error message FIRST - specific patterns take priority over status codes
if (errorText) {
const errorStr = typeof errorText === "string" ? errorText : JSON.stringify(errorText);
const lowerError = errorStr.toLowerCase();
const lowerError = errorText
? (typeof errorText === "string" ? errorText : JSON.stringify(errorText)).toLowerCase()
: "";
if (lowerError.includes("no credentials")) {
return { shouldFallback: true, cooldownMs: COOLDOWN_MS.notFound };
for (const rule of ERROR_RULES) {
// Text-based rule: match substring in error message
if (rule.text && lowerError && lowerError.includes(rule.text)) {
if (rule.backoff) {
const newLevel = Math.min(backoffLevel + 1, BACKOFF_CONFIG.maxLevel);
return { shouldFallback: true, cooldownMs: getQuotaCooldown(backoffLevel), newBackoffLevel: newLevel };
}
return { shouldFallback: true, cooldownMs: rule.cooldownMs };
}
if (lowerError.includes("request not allowed")) {
return { shouldFallback: true, cooldownMs: COOLDOWN_MS.requestNotAllowed };
}
// Kiro: "improperly formed request" = model not available on this account tier
// Treat as paymentRequired (long cooldown) so the model is locked and fallback occurs
if (lowerError.includes("improperly formed request")) {
return { shouldFallback: true, cooldownMs: COOLDOWN_MS.paymentRequired };
}
// Rate limit keywords - exponential backoff
if (
lowerError.includes("rate limit") ||
lowerError.includes("too many requests") ||
lowerError.includes("quota exceeded") ||
lowerError.includes("capacity") ||
lowerError.includes("overloaded")
) {
const newLevel = Math.min(backoffLevel + 1, BACKOFF_CONFIG.maxLevel);
return {
shouldFallback: true,
cooldownMs: getQuotaCooldown(backoffLevel),
newBackoffLevel: newLevel
};
// Status-based rule: match HTTP status code
if (rule.status && rule.status === status) {
if (rule.backoff) {
const newLevel = Math.min(backoffLevel + 1, BACKOFF_CONFIG.maxLevel);
return { shouldFallback: true, cooldownMs: getQuotaCooldown(backoffLevel), newBackoffLevel: newLevel };
}
return { shouldFallback: true, cooldownMs: rule.cooldownMs };
}
}
if (status === HTTP_STATUS.UNAUTHORIZED) {
return { shouldFallback: true, cooldownMs: COOLDOWN_MS.unauthorized };
}
if (status === HTTP_STATUS.PAYMENT_REQUIRED || status === HTTP_STATUS.FORBIDDEN) {
return { shouldFallback: true, cooldownMs: COOLDOWN_MS.paymentRequired };
}
if (status === HTTP_STATUS.NOT_FOUND) {
return { shouldFallback: true, cooldownMs: COOLDOWN_MS.notFound };
}
// 429 - Rate limit with exponential backoff
if (status === HTTP_STATUS.RATE_LIMITED) {
const newLevel = Math.min(backoffLevel + 1, BACKOFF_CONFIG.maxLevel);
return {
shouldFallback: true,
cooldownMs: getQuotaCooldown(backoffLevel),
newBackoffLevel: newLevel
};
}
// Transient errors
const transientStatuses = [
HTTP_STATUS.NOT_ACCEPTABLE, HTTP_STATUS.REQUEST_TIMEOUT,
HTTP_STATUS.SERVER_ERROR, HTTP_STATUS.BAD_GATEWAY,
HTTP_STATUS.SERVICE_UNAVAILABLE, HTTP_STATUS.GATEWAY_TIMEOUT
];
if (transientStatuses.includes(status)) {
return { shouldFallback: true, cooldownMs: COOLDOWN_MS.transient };
}
// All other errors - fallback with transient cooldown
return { shouldFallback: true, cooldownMs: COOLDOWN_MS.transient };
// Default: transient cooldown for any unmatched error
return { shouldFallback: true, cooldownMs: TRANSIENT_COOLDOWN_MS };
}
/**

View file

@ -1,4 +1,4 @@
import { ERROR_TYPES, DEFAULT_ERROR_MESSAGES } from "../config/runtimeConfig.js";
import { ERROR_TYPES, DEFAULT_ERROR_MESSAGES } from "../config/errorConfig.js";
/**
* Build OpenAI-compatible error response body
@ -49,56 +49,17 @@ export async function writeStreamError(writer, statusCode, message) {
await writer.write(encoder.encode(`data: ${JSON.stringify(errorBody)}\n\n`));
}
/**
* Parse Antigravity error message to extract retry time
* Example: "You have exhausted your capacity on this model. Your quota will reset after 2h7m23s."
* @param {string} message - Error message
* @returns {number|null} Retry time in milliseconds, or null if not found
*/
export function parseAntigravityRetryTime(message) {
if (typeof message !== "string") return null;
// Match patterns like: 2h7m23s, 5m30s, 45s, 1h20m, etc.
const match = message.match(/reset after (\d+h)?(\d+m)?(\d+s)?/i);
if (!match) return null;
let totalMs = 0;
// Extract hours
if (match[1]) {
const hours = parseInt(match[1]);
totalMs += hours * 60 * 60 * 1000;
}
// Extract minutes
if (match[2]) {
const minutes = parseInt(match[2]);
totalMs += minutes * 60 * 1000;
}
// Extract seconds
if (match[3]) {
const seconds = parseInt(match[3]);
totalMs += seconds * 1000;
}
return totalMs > 0 ? totalMs : null;
}
/**
* Parse upstream provider error response
* @param {Response} response - Fetch response from provider
* @param {string} provider - Provider name (for Antigravity-specific parsing)
* @returns {Promise<{statusCode: number, message: string, retryAfterMs: number|null}>}
* @returns {Promise<{statusCode: number, message: string}>}
*/
export async function parseUpstreamError(response, provider = null) {
export async function parseUpstreamError(response) {
let message = "";
let retryAfterMs = null;
try {
const text = await response.text();
// Try parse as JSON
try {
const json = JSON.parse(text);
message = json.error?.message || json.message || json.error || text;
@ -112,15 +73,9 @@ export async function parseUpstreamError(response, provider = null) {
const messageStr = typeof message === "string" ? message : JSON.stringify(message);
const finalMessage = messageStr || DEFAULT_ERROR_MESSAGES[response.status] || `Upstream error: ${response.status}`;
// Parse Antigravity-specific retry time from error message
if (provider === "antigravity" && response.status === 429) {
retryAfterMs = parseAntigravityRetryTime(finalMessage);
}
return {
statusCode: response.status,
message: finalMessage,
retryAfterMs
message: finalMessage
};
}
@ -128,23 +83,15 @@ export async function parseUpstreamError(response, provider = null) {
* Create error result for chatCore handler
* @param {number} statusCode - HTTP status code
* @param {string} message - Error message
* @param {number|null} retryAfterMs - Optional retry-after time in milliseconds
* @returns {{ success: false, status: number, error: string, response: Response, retryAfterMs?: number }}
* @returns {{ success: false, status: number, error: string, response: Response }}
*/
export function createErrorResult(statusCode, message, retryAfterMs = null) {
const result = {
export function createErrorResult(statusCode, message) {
return {
success: false,
status: statusCode,
error: message,
response: errorResponse(statusCode, message)
};
// Add retryAfterMs if available (for Antigravity quota errors)
if (retryAfterMs) {
result.retryAfterMs = retryAfterMs;
}
return result;
}
/**

View file

@ -11,6 +11,7 @@ import { useCopyToClipboard } from "@/shared/hooks/useCopyToClipboard";
import ConnectionsCard from "@/app/(dashboard)/dashboard/providers/components/ConnectionsCard";
import ModelsCard from "@/app/(dashboard)/dashboard/providers/components/ModelsCard";
import { TTS_PROVIDER_CONFIG } from "@/shared/constants/ttsProviders";
import { getTtsVoicesForModel } from "open-sse/config/ttsModels.js";
// Shared row layout — defined outside components to avoid re-mount on re-render
function Row({ label, children }) {
@ -40,6 +41,60 @@ const DEFAULT_RESPONSE_EXAMPLE = `{
"usage": { "prompt_tokens": 9, "total_tokens": 9 }
}`;
// Config-driven example defaults per kind
const KIND_EXAMPLE_CONFIG = {
webSearch: {
inputLabel: "Query",
inputPlaceholder: "What is the latest news about AI?",
defaultInput: "What is the latest news about AI?",
bodyKey: "query",
defaultResponse: `{\n "results": [\n { "title": "...", "url": "...", "snippet": "..." }\n ]\n}`,
},
webFetch: {
inputLabel: "URL",
inputPlaceholder: "https://example.com",
defaultInput: "https://example.com",
bodyKey: "url",
defaultResponse: `{\n "content": "...",\n "title": "...",\n "url": "..."\n}`,
},
image: {
inputLabel: "Prompt",
inputPlaceholder: "A cute cat wearing a hat",
defaultInput: "A cute cat wearing a hat",
bodyKey: "prompt",
defaultResponse: `{\n "data": [\n { "url": "...", "b64_json": "..." }\n ]\n}`,
},
imageToText: {
inputLabel: "Image URL",
inputPlaceholder: "https://example.com/image.png",
defaultInput: "https://upload.wikimedia.org/wikipedia/commons/thumb/3/3a/Cat03.jpg/1200px-Cat03.jpg",
bodyKey: "url",
extraBody: { prompt: "Describe this image in detail" },
defaultResponse: `{\n "text": "A cat sitting on a windowsill...",\n "model": "..."\n}`,
},
stt: {
inputLabel: "Audio URL",
inputPlaceholder: "https://example.com/audio.mp3",
defaultInput: "",
bodyKey: "url",
defaultResponse: `{\n "text": "Hello world...",\n "model": "..."\n}`,
},
video: {
inputLabel: "Prompt",
inputPlaceholder: "A serene lake at sunset",
defaultInput: "A serene lake at sunset",
bodyKey: "prompt",
defaultResponse: `{\n "data": [\n { "url": "..." }\n ]\n}`,
},
music: {
inputLabel: "Prompt",
inputPlaceholder: "A calm piano melody",
defaultInput: "A calm piano melody",
bodyKey: "prompt",
defaultResponse: `{\n "data": [\n { "url": "...", "format": "mp3" }\n ]\n}`,
},
};
// EmbeddingExampleCard
function EmbeddingExampleCard({ providerId }) {
const providerAlias = getProviderAlias(providerId);
@ -300,8 +355,13 @@ function TtsExampleCard({ providerId }) {
// Pre-select default voice based on provider config
if (config.voiceSource === "hardcoded") {
const voiceKey = config.voiceKey || providerId;
const voices = getModelsByProviderId(voiceKey).filter((m) => m.type === "tts");
const defaultModel = config.hasModelSelector && config.modelKey
? (getModelsByProviderId(config.modelKey)?.[0]?.id || "")
: "";
// Use per-model voices if available, else flat list
const voices = (config.voicesPerModel && defaultModel)
? (getTtsVoicesForModel(providerId, defaultModel) || [])
: getModelsByProviderId(config.voiceKey || providerId).filter((m) => m.type === "tts");
if (voices.length) {
if (config.hasBrowseButton) {
// Google TTS: pre-select "en" (English) as default, show as single voice chip
@ -311,7 +371,7 @@ function TtsExampleCard({ providerId }) {
setSelectedVoiceName(defaultVoice.name);
setCountryVoices([{ id: defaultVoice.id, name: defaultVoice.name }]);
} else {
// OpenAI: set voice chips directly (no language picker)
// OpenAI/OpenRouter: set voice chips directly (no language picker)
setCountryVoices(voices);
setSelectedVoice(voices[0].id);
setSelectedVoiceName(voices[0].name || voices[0].id);
@ -321,6 +381,17 @@ function TtsExampleCard({ providerId }) {
// api-language (edge-tts, local-device, elevenlabs): NO default load, wait for user to pick language
}, [providerId]);
// Update voices when model changes (voicesPerModel providers)
useEffect(() => {
if (!config.voicesPerModel || !selectedModel) return;
const voices = getTtsVoicesForModel(providerId, selectedModel) || [];
setCountryVoices(voices);
if (voices.length) {
setSelectedVoice(voices[0].id);
setSelectedVoiceName(voices[0].name || voices[0].id);
}
}, [selectedModel]);
// Open modal — load language list
const openModal = async () => {
setModalOpen(true);
@ -745,6 +816,186 @@ function TtsExampleCard({ providerId }) {
);
}
// Generic Example Card — config-driven for webSearch, webFetch, image, imageToText, stt, video, music
function GenericExampleCard({ providerId, kind }) {
const providerAlias = getProviderAlias(providerId);
const kindConfig = MEDIA_PROVIDER_KINDS.find((k) => k.id === kind);
const exConfig = KIND_EXAMPLE_CONFIG[kind];
if (!kindConfig || !exConfig) return null;
const [input, setInput] = useState(exConfig.defaultInput);
const [apiKey, setApiKey] = useState("");
const [useTunnel, setUseTunnel] = useState(false);
const [localEndpoint, setLocalEndpoint] = useState("");
const [tunnelEndpoint, setTunnelEndpoint] = useState("");
const [result, setResult] = useState(null);
const [running, setRunning] = useState(false);
const [error, setError] = useState("");
const { copied: copiedCurl, copy: copyCurl } = useCopyToClipboard();
const { copied: copiedRes, copy: copyRes } = useCopyToClipboard();
useEffect(() => {
setLocalEndpoint(window.location.origin);
fetch("/api/keys")
.then((r) => r.json())
.then((d) => { setApiKey((d.keys || []).find((k) => k.isActive !== false)?.key || ""); })
.catch(() => {});
fetch("/api/tunnel/status")
.then((r) => r.json())
.then((d) => { if (d.publicUrl) setTunnelEndpoint(d.publicUrl); })
.catch(() => {});
}, []);
const endpoint = useTunnel ? tunnelEndpoint : localEndpoint;
const apiPath = kindConfig.endpoint.path;
const requestBody = {
model: `${providerAlias}/model-name`,
[exConfig.bodyKey]: input,
...exConfig.extraBody,
};
const curlSnippet = `curl -X ${kindConfig.endpoint.method} ${endpoint}${apiPath} \\
-H "Content-Type: application/json" \\
-H "Authorization: Bearer ${apiKey || "YOUR_KEY"}" \\
-d '${JSON.stringify(requestBody)}'`;
const handleRun = async () => {
if (!input.trim()) return;
setRunning(true);
setError("");
setResult(null);
const start = Date.now();
try {
const headers = { "Content-Type": "application/json" };
if (apiKey) headers["Authorization"] = `Bearer ${apiKey}`;
const body = { ...requestBody, model: `${providerAlias}/model-name` };
const res = await fetch(`/api${apiPath}`, {
method: kindConfig.endpoint.method,
headers,
body: JSON.stringify(body),
});
const latencyMs = Date.now() - start;
const data = await res.json();
if (!res.ok) { setError(data?.error?.message || data?.error || `HTTP ${res.status}`); return; }
setResult({ data, latencyMs });
} catch (e) {
setError(e.message || "Network error");
} finally {
setRunning(false);
}
};
const resultJson = result ? JSON.stringify(result.data, null, 2) : "";
return (
<Card>
<h2 className="text-lg font-semibold mb-4">Example</h2>
<div className="flex flex-col gap-2.5">
{/* Endpoint */}
<Row label="Endpoint">
<div className="flex items-center gap-2">
<span className="flex-1 px-3 py-1.5 text-sm font-mono text-text-main bg-sidebar rounded-lg truncate">
{endpoint}{apiPath}
</span>
{tunnelEndpoint && (
<button
onClick={() => setUseTunnel((v) => !v)}
title={useTunnel ? "Using tunnel" : "Using local"}
className={`flex items-center gap-1 text-xs px-2 py-1.5 rounded-lg border shrink-0 transition-colors ${
useTunnel ? "border-primary/40 bg-primary/10 text-primary" : "border-border text-text-muted hover:text-primary"
}`}
>
<span className="material-symbols-outlined text-[14px]">wifi_tethering</span>
Tunnel
</button>
)}
</div>
</Row>
{/* API Key */}
<Row label="API Key">
<span className="px-3 py-1.5 text-sm font-mono text-text-main bg-sidebar rounded-lg truncate block">
{apiKey ? `${apiKey.slice(0, 8)}${"\u2022".repeat(Math.min(20, apiKey.length - 8))}` : <span className="text-text-muted italic">No key configured</span>}
</span>
</Row>
{/* Input */}
<Row label={exConfig.inputLabel}>
<div className="relative">
<input
value={input}
onChange={(e) => setInput(e.target.value)}
placeholder={exConfig.inputPlaceholder}
className="w-full px-3 py-1.5 pr-7 text-sm border border-border rounded-lg bg-background focus:outline-none focus:border-primary"
/>
{input && (
<button
type="button"
onClick={() => setInput("")}
className="absolute right-2 top-1/2 -translate-y-1/2 text-text-muted hover:text-primary transition-colors"
>
<span className="material-symbols-outlined text-[14px]">close</span>
</button>
)}
</div>
</Row>
{/* Curl + Run */}
<div className="mt-1">
<div className="flex items-center justify-between mb-1.5">
<span className="text-xs font-semibold text-text-muted uppercase tracking-wider">Request</span>
<div className="flex items-center gap-2">
<button
onClick={() => copyCurl(curlSnippet)}
className="flex items-center gap-1 text-xs text-text-muted hover:text-primary transition-colors"
>
<span className="material-symbols-outlined text-[14px]">{copiedCurl ? "check" : "content_copy"}</span>
{copiedCurl ? "Copied" : "Copy"}
</button>
<button
onClick={handleRun}
disabled={running || !input.trim()}
className="flex items-center gap-1.5 px-3 py-1 rounded-lg bg-primary text-white text-xs font-medium hover:bg-primary/90 transition-colors disabled:opacity-50 disabled:cursor-not-allowed"
>
<span className="material-symbols-outlined text-[14px]" style={running ? { animation: "spin 1s linear infinite" } : undefined}>
{running ? "progress_activity" : "play_arrow"}
</span>
{running ? "Running..." : "Run"}
</button>
</div>
</div>
<pre className="bg-sidebar rounded-lg px-3 py-2.5 text-xs font-mono text-text-main overflow-x-auto whitespace-pre">{curlSnippet}</pre>
</div>
{/* Error */}
{error && <p className="text-xs text-red-500 break-words">{error}</p>}
{/* Response */}
<div>
<div className="flex items-center justify-between mb-1.5">
<span className="text-xs font-semibold text-text-muted uppercase tracking-wider">
Response {result && <span className="font-normal normal-case">&#9889; {result.latencyMs}ms</span>}
</span>
{result && (
<button
onClick={() => copyRes(resultJson)}
className="flex items-center gap-1 text-xs text-text-muted hover:text-primary transition-colors"
>
<span className="material-symbols-outlined text-[14px]">{copiedRes ? "check" : "content_copy"}</span>
{copiedRes ? "Copied" : "Copy"}
</button>
)}
</div>
<pre className="bg-sidebar rounded-lg px-3 py-2.5 text-xs font-mono text-text-main overflow-x-auto whitespace-pre opacity-70">
{result ? resultJson : exConfig.defaultResponse}
</pre>
</div>
</div>
</Card>
);
}
// MediaProviderDetailPage
export default function MediaProviderDetailPage() {
const { kind, id } = useParams();
@ -817,6 +1068,7 @@ export default function MediaProviderDetailPage() {
{/* Example — per kind */}
{kind === "embedding" && <EmbeddingExampleCard providerId={id} />}
{kind === "tts" && <TtsExampleCard providerId={id} />}
{KIND_EXAMPLE_CONFIG[kind] && <GenericExampleCard providerId={id} kind={kind} />}
</div>
);
}

View file

@ -665,8 +665,9 @@ export default function ProviderDetailPage() {
{/* Suggested models from provider API — show only models not yet added */}
{suggestedModels.length > 0 && (() => {
const addedFullModels = new Set(Object.values(modelAliases));
const hardcodedIds = new Set(models.map((m) => m.id));
const notAdded = suggestedModels.filter(
(m) => !addedFullModels.has(`${providerStorageAlias}/${m.id}`)
(m) => !addedFullModels.has(`${providerStorageAlias}/${m.id}`) && !hardcodedIds.has(m.id)
);
if (notAdded.length === 0) return null;
return (

View file

@ -0,0 +1,49 @@
import { NextResponse } from "next/server";
export const dynamic = "force-dynamic";
const FILTERS = {
"openrouter-free": (models) =>
models
.filter(
(m) =>
m.pricing?.prompt === "0" &&
m.pricing?.completion === "0" &&
m.context_length >= 200000
)
.map((m) => ({ id: m.id, name: m.name, contextLength: m.context_length }))
.sort((a, b) => b.contextLength - a.contextLength),
"opencode-free": (models) =>
models
.filter((m) => m.id?.endsWith("-free"))
.map((m) => ({ id: m.id, name: m.id })),
};
export async function GET(request) {
const { searchParams } = new URL(request.url);
const url = searchParams.get("url");
const type = searchParams.get("type");
if (!url || !type) {
return NextResponse.json({ error: "Missing url or type" }, { status: 400 });
}
const filter = FILTERS[type];
if (!filter) {
return NextResponse.json({ error: "Unknown filter type" }, { status: 400 });
}
try {
const res = await fetch(url);
if (!res.ok) {
return NextResponse.json({ data: [] });
}
const json = await res.json();
const raw = json.data ?? json.models ?? json;
const data = filter(Array.isArray(raw) ? raw : []);
return NextResponse.json({ data });
} catch {
return NextResponse.json({ data: [] });
}
}

File diff suppressed because it is too large Load diff

View file

@ -64,12 +64,74 @@ if (!isCloud && fs && typeof fs.existsSync === "function") {
}
}
// Default data structure
const defaultData = {
history: [],
totalRequestsLifetime: 0
totalRequestsLifetime: 0,
dailySummary: {},
};
function getLocalDateKey(timestamp) {
const d = timestamp ? new Date(timestamp) : new Date();
return `${d.getFullYear()}-${String(d.getMonth() + 1).padStart(2, "0")}-${String(d.getDate()).padStart(2, "0")}`;
}
function addToCounter(target, key, values) {
if (!target[key]) target[key] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0 };
target[key].requests += values.requests || 1;
target[key].promptTokens += values.promptTokens || 0;
target[key].completionTokens += values.completionTokens || 0;
target[key].cost += values.cost || 0;
if (values.meta) Object.assign(target[key], values.meta);
}
function aggregateEntryToDailySummary(dailySummary, entry) {
const dateKey = getLocalDateKey(entry.timestamp);
if (!dailySummary[dateKey]) {
dailySummary[dateKey] = {
requests: 0, promptTokens: 0, completionTokens: 0, cost: 0,
byProvider: {}, byModel: {}, byAccount: {}, byApiKey: {}, byEndpoint: {},
};
}
const day = dailySummary[dateKey];
const promptTokens = entry.tokens?.prompt_tokens || entry.tokens?.input_tokens || 0;
const completionTokens = entry.tokens?.completion_tokens || entry.tokens?.output_tokens || 0;
const cost = entry.cost || 0;
const vals = { promptTokens, completionTokens, cost };
day.requests += 1;
day.promptTokens += promptTokens;
day.completionTokens += completionTokens;
day.cost += cost;
if (entry.provider) addToCounter(day.byProvider, entry.provider, vals);
const modelKey = entry.provider ? `${entry.model}|${entry.provider}` : entry.model;
addToCounter(day.byModel, modelKey, { ...vals, meta: { rawModel: entry.model, provider: entry.provider } });
if (entry.connectionId) {
addToCounter(day.byAccount, entry.connectionId, { ...vals, meta: { rawModel: entry.model, provider: entry.provider } });
}
const apiKeyVal = entry.apiKey && typeof entry.apiKey === "string" ? entry.apiKey : "local-no-key";
const akModelKey = `${apiKeyVal}|${entry.model}|${entry.provider || "unknown"}`;
addToCounter(day.byApiKey, akModelKey, { ...vals, meta: { rawModel: entry.model, provider: entry.provider, apiKey: entry.apiKey || null } });
const endpoint = entry.endpoint || "Unknown";
const epKey = `${endpoint}|${entry.model}|${entry.provider || "unknown"}`;
addToCounter(day.byEndpoint, epKey, { ...vals, meta: { endpoint, rawModel: entry.model, provider: entry.provider } });
}
function migrateHistoryToDailySummary(db) {
const history = db.data.history || [];
if (!history.length) return false;
db.data.dailySummary = {};
for (const entry of history) {
aggregateEntryToDailySummary(db.data.dailySummary, entry);
}
console.log(`[usageDb] Migrated ${history.length} history entries to dailySummary (${Object.keys(db.data.dailySummary).length} days)`);
return true;
}
// Singleton instance
let dbInstance = null;
@ -238,11 +300,19 @@ export async function getUsageDb() {
}
}
// Initialize with default data if empty
if (!dbInstance.data) {
dbInstance.data = defaultData;
dbInstance.data = { ...defaultData };
await dbInstance.write();
}
// Migration: build dailySummary from existing history (one-time)
if (!dbInstance.data.dailySummary) {
if (migrateHistoryToDailySummary(dbInstance)) {
await dbInstance.write();
} else {
dbInstance.data.dailySummary = {};
}
}
}
return dbInstance;
}
@ -275,7 +345,9 @@ export async function saveRequestUsage(entry) {
db.data.history.push(entry);
db.data.totalRequestsLifetime += 1;
// Cap history to prevent unbounded memory/disk growth
if (!db.data.dailySummary) db.data.dailySummary = {};
aggregateEntryToDailySummary(db.data.dailySummary, entry);
const MAX_HISTORY = 10000;
if (db.data.history.length > MAX_HISTORY) {
db.data.history.splice(0, db.data.history.length - MAX_HISTORY);
@ -470,33 +542,18 @@ const PERIOD_MS = { "24h": 86400000, "7d": 604800000, "30d": 2592000000, "60d":
*/
export async function getUsageStats(period = "all") {
const db = await getUsageDb();
let history = db.data.history || [];
const history = db.data.history || [];
const dailySummary = db.data.dailySummary || {};
// Filter history by period
if (period && PERIOD_MS[period]) {
const cutoff = Date.now() - PERIOD_MS[period];
history = history.filter((e) => new Date(e.timestamp).getTime() >= cutoff);
}
// Import localDb to get provider connection names and API keys
const { getProviderConnections, getApiKeys, getProviderNodes } = await import("@/lib/localDb.js");
// Fetch all provider connections to get account names
let allConnections = [];
try {
allConnections = await getProviderConnections();
} catch (error) {
// If localDb is not available (e.g., in some environments), continue without account names
console.warn("Could not fetch provider connections for usage stats:", error.message);
}
// Create a map from connectionId to account name
try { allConnections = await getProviderConnections(); } catch {}
const connectionMap = {};
for (const conn of allConnections) {
connectionMap[conn.id] = conn.name || conn.email || conn.id;
}
// Build map from compatible provider ID → friendly name (from providerNodes)
const providerNodeNameMap = {};
try {
const nodes = await getProviderNodes();
@ -505,44 +562,28 @@ export async function getUsageStats(period = "all") {
}
} catch {}
// Fetch all API keys to get key names
let allApiKeys = [];
try {
allApiKeys = await getApiKeys();
} catch (error) {
console.warn("Could not fetch API keys for usage stats:", error.message);
}
// Create a map from API key to key info
try { allApiKeys = await getApiKeys(); } catch {}
const apiKeyMap = {};
for (const key of allApiKeys) {
apiKeyMap[key.key] = {
name: key.name,
id: key.id,
createdAt: key.createdAt
};
apiKeyMap[key.key] = { name: key.name, id: key.id, createdAt: key.createdAt };
}
// 20 most recent requests from history (always in sync with SSE emit)
// Recent requests (always from live history)
const seen = new Set();
const recentRequests = [...history]
.sort((a, b) => new Date(b.timestamp) - new Date(a.timestamp))
.map((e) => {
const t = e.tokens || {};
const promptTokens = t.prompt_tokens || t.input_tokens || 0;
const completionTokens = t.completion_tokens || t.output_tokens || 0;
return {
timestamp: e.timestamp,
model: e.model,
provider: e.provider || "",
promptTokens,
completionTokens,
timestamp: e.timestamp, model: e.model, provider: e.provider || "",
promptTokens: t.prompt_tokens || t.input_tokens || 0,
completionTokens: t.completion_tokens || t.output_tokens || 0,
status: e.status || "ok",
};
})
.filter((e) => {
if (e.promptTokens === 0 && e.completionTokens === 0) return false;
// Deduplicate: same model+provider+tokens within same minute
const minute = e.timestamp ? e.timestamp.slice(0, 16) : "";
const key = `${e.model}|${e.provider}|${e.promptTokens}|${e.completionTokens}|${minute}`;
if (seen.has(key)) return false;
@ -557,14 +598,8 @@ export async function getUsageStats(period = "all") {
const stats = {
totalRequests: lifetimeTotalRequests,
totalPromptTokens: 0,
totalCompletionTokens: 0,
totalCost: 0,
byProvider: {},
byModel: {},
byAccount: {},
byApiKey: {},
byEndpoint: {},
totalPromptTokens: 0, totalCompletionTokens: 0, totalCost: 0,
byProvider: {}, byModel: {}, byAccount: {}, byApiKey: {}, byEndpoint: {},
last10Minutes: [],
pending: pendingRequests,
activeRequests: [],
@ -572,218 +607,221 @@ export async function getUsageStats(period = "all") {
errorProvider: (Date.now() - lastErrorProvider.ts < 10000) ? lastErrorProvider.provider : "",
};
// Build active requests list from pending counts
// Active requests from pending
for (const [connectionId, models] of Object.entries(pendingRequests.byAccount)) {
for (const [modelKey, count] of Object.entries(models)) {
if (count > 0) {
const accountName = connectionMap[connectionId] || `Account ${connectionId.slice(0, 8)}...`;
// modelKey is "model (provider)"
const match = modelKey.match(/^(.*) \((.*)\)$/);
const modelName = match ? match[1] : modelKey;
const providerName = match ? match[2] : "unknown";
stats.activeRequests.push({
model: modelName,
provider: providerName,
account: accountName,
count
model: match ? match[1] : modelKey,
provider: match ? match[2] : "unknown",
account: accountName, count,
});
}
}
}
// Initialize 10-minute buckets using stable minute boundaries
// last10Minutes — always from live history
const now = new Date();
// Floor to the start of the current minute
const currentMinuteStart = new Date(Math.floor(now.getTime() / 60000) * 60000);
const tenMinutesAgo = new Date(currentMinuteStart.getTime() - 9 * 60 * 1000);
// Create buckets keyed by minute timestamp for stable lookups
const bucketMap = {};
for (let i = 0; i < 10; i++) {
const bucketTime = new Date(currentMinuteStart.getTime() - (9 - i) * 60 * 1000);
const bucketKey = bucketTime.getTime();
bucketMap[bucketKey] = {
requests: 0,
promptTokens: 0,
completionTokens: 0,
cost: 0
};
const bucketKey = currentMinuteStart.getTime() - (9 - i) * 60 * 1000;
bucketMap[bucketKey] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0 };
stats.last10Minutes.push(bucketMap[bucketKey]);
}
for (const entry of history) {
const promptTokens = entry.tokens?.prompt_tokens || 0;
const completionTokens = entry.tokens?.completion_tokens || 0;
const entryTime = new Date(entry.timestamp);
// Use pre-stored cost (saved at request time), avoid recalculating
const entryCost = entry.cost || 0;
stats.totalPromptTokens += promptTokens;
stats.totalCompletionTokens += completionTokens;
stats.totalCost += entryCost;
// Last 10 minutes aggregation - floor entry time to its minute
if (entryTime >= tenMinutesAgo && entryTime <= now) {
const entryMinuteStart = Math.floor(entryTime.getTime() / 60000) * 60000;
if (bucketMap[entryMinuteStart]) {
const pt = entry.tokens?.prompt_tokens || 0;
const ct = entry.tokens?.completion_tokens || 0;
bucketMap[entryMinuteStart].requests++;
bucketMap[entryMinuteStart].promptTokens += promptTokens;
bucketMap[entryMinuteStart].completionTokens += completionTokens;
bucketMap[entryMinuteStart].cost += entryCost;
bucketMap[entryMinuteStart].promptTokens += pt;
bucketMap[entryMinuteStart].completionTokens += ct;
bucketMap[entryMinuteStart].cost += entry.cost || 0;
}
}
}
// By Provider
if (!stats.byProvider[entry.provider]) {
stats.byProvider[entry.provider] = {
requests: 0,
promptTokens: 0,
completionTokens: 0,
cost: 0
};
}
stats.byProvider[entry.provider].requests++;
stats.byProvider[entry.provider].promptTokens += promptTokens;
stats.byProvider[entry.provider].completionTokens += completionTokens;
stats.byProvider[entry.provider].cost += entryCost;
// Determine if we use dailySummary (7d/30d/60d/all) or live history (24h)
const useDailySummary = period !== "24h";
// By Model
// Format: "modelName (provider)" if provider is known
const modelKey = entry.provider ? `${entry.model} (${entry.provider})` : entry.model;
// Resolve friendly name for compatible providers
const providerDisplayName = providerNodeNameMap[entry.provider] || entry.provider;
if (useDailySummary) {
// Collect relevant date keys
const periodDays = { "7d": 7, "30d": 30, "60d": 60 };
const maxDays = periodDays[period] || null; // null = all
const today = new Date();
const dateKeys = Object.keys(dailySummary).filter((dateKey) => {
if (!maxDays) return true;
const parts = dateKey.split("-");
const d = new Date(Number(parts[0]), Number(parts[1]) - 1, Number(parts[2]));
const diffDays = Math.floor((today.getTime() - d.getTime()) / 86400000);
return diffDays < maxDays;
});
if (!stats.byModel[modelKey]) {
stats.byModel[modelKey] = {
requests: 0,
promptTokens: 0,
completionTokens: 0,
cost: 0,
rawModel: entry.model,
provider: providerDisplayName,
lastUsed: entry.timestamp
};
}
stats.byModel[modelKey].requests++;
stats.byModel[modelKey].promptTokens += promptTokens;
stats.byModel[modelKey].completionTokens += completionTokens;
stats.byModel[modelKey].cost += entryCost;
if (new Date(entry.timestamp) > new Date(stats.byModel[modelKey].lastUsed)) {
stats.byModel[modelKey].lastUsed = entry.timestamp;
}
for (const dateKey of dateKeys) {
const day = dailySummary[dateKey];
stats.totalPromptTokens += day.promptTokens || 0;
stats.totalCompletionTokens += day.completionTokens || 0;
stats.totalCost += day.cost || 0;
// By Account (model + oauth account)
// Use connectionId if available, otherwise fallback to provider name
if (entry.connectionId) {
const accountName = connectionMap[entry.connectionId] || `Account ${entry.connectionId.slice(0, 8)}...`;
const accountKey = `${entry.model} (${entry.provider} - ${accountName})`;
if (!stats.byAccount[accountKey]) {
stats.byAccount[accountKey] = {
requests: 0,
promptTokens: 0,
completionTokens: 0,
cost: 0,
rawModel: entry.model,
provider: providerDisplayName,
connectionId: entry.connectionId,
accountName: accountName,
lastUsed: entry.timestamp
};
// Merge byProvider
for (const [prov, pData] of Object.entries(day.byProvider || {})) {
if (!stats.byProvider[prov]) stats.byProvider[prov] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0 };
stats.byProvider[prov].requests += pData.requests || 0;
stats.byProvider[prov].promptTokens += pData.promptTokens || 0;
stats.byProvider[prov].completionTokens += pData.completionTokens || 0;
stats.byProvider[prov].cost += pData.cost || 0;
}
stats.byAccount[accountKey].requests++;
stats.byAccount[accountKey].promptTokens += promptTokens;
stats.byAccount[accountKey].completionTokens += completionTokens;
stats.byAccount[accountKey].cost += entryCost;
if (new Date(entry.timestamp) > new Date(stats.byAccount[accountKey].lastUsed)) {
stats.byAccount[accountKey].lastUsed = entry.timestamp;
// Merge byModel (dailySummary key: "model|provider" → stats key: "model (provider)")
for (const [mk, mData] of Object.entries(day.byModel || {})) {
const rawModel = mData.rawModel || mk.split("|")[0];
const provider = mData.provider || mk.split("|")[1] || "";
const statsKey = provider ? `${rawModel} (${provider})` : rawModel;
const providerDisplayName = providerNodeNameMap[provider] || provider;
if (!stats.byModel[statsKey]) {
stats.byModel[statsKey] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0, rawModel, provider: providerDisplayName, lastUsed: dateKey };
}
stats.byModel[statsKey].requests += mData.requests || 0;
stats.byModel[statsKey].promptTokens += mData.promptTokens || 0;
stats.byModel[statsKey].completionTokens += mData.completionTokens || 0;
stats.byModel[statsKey].cost += mData.cost || 0;
if (dateKey > (stats.byModel[statsKey].lastUsed || "")) stats.byModel[statsKey].lastUsed = dateKey;
}
// Merge byAccount
for (const [connId, aData] of Object.entries(day.byAccount || {})) {
const accountName = connectionMap[connId] || `Account ${connId.slice(0, 8)}...`;
const rawModel = aData.rawModel || "";
const provider = aData.provider || "";
const providerDisplayName = providerNodeNameMap[provider] || provider;
const accountKey = `${rawModel} (${provider} - ${accountName})`;
if (!stats.byAccount[accountKey]) {
stats.byAccount[accountKey] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0, rawModel, provider: providerDisplayName, connectionId: connId, accountName, lastUsed: dateKey };
}
stats.byAccount[accountKey].requests += aData.requests || 0;
stats.byAccount[accountKey].promptTokens += aData.promptTokens || 0;
stats.byAccount[accountKey].completionTokens += aData.completionTokens || 0;
stats.byAccount[accountKey].cost += aData.cost || 0;
if (dateKey > (stats.byAccount[accountKey].lastUsed || "")) stats.byAccount[accountKey].lastUsed = dateKey;
}
// Merge byApiKey
for (const [akKey, akData] of Object.entries(day.byApiKey || {})) {
const rawModel = akData.rawModel || "";
const provider = akData.provider || "";
const providerDisplayName = providerNodeNameMap[provider] || provider;
const apiKeyVal = akData.apiKey;
const keyInfo = apiKeyVal ? apiKeyMap[apiKeyVal] : null;
const keyName = keyInfo?.name || (apiKeyVal ? apiKeyVal.slice(0, 8) + "..." : "Local (No API Key)");
const apiKeyKey = apiKeyVal || "local-no-key";
if (!stats.byApiKey[akKey]) {
stats.byApiKey[akKey] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0, rawModel, provider: providerDisplayName, apiKey: apiKeyVal, keyName, apiKeyKey, lastUsed: dateKey };
}
stats.byApiKey[akKey].requests += akData.requests || 0;
stats.byApiKey[akKey].promptTokens += akData.promptTokens || 0;
stats.byApiKey[akKey].completionTokens += akData.completionTokens || 0;
stats.byApiKey[akKey].cost += akData.cost || 0;
if (dateKey > (stats.byApiKey[akKey].lastUsed || "")) stats.byApiKey[akKey].lastUsed = dateKey;
}
// Merge byEndpoint
for (const [epKey, epData] of Object.entries(day.byEndpoint || {})) {
const endpoint = epData.endpoint || epKey.split("|")[0] || "Unknown";
const rawModel = epData.rawModel || "";
const provider = epData.provider || "";
const providerDisplayName = providerNodeNameMap[provider] || provider;
if (!stats.byEndpoint[epKey]) {
stats.byEndpoint[epKey] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0, endpoint, rawModel, provider: providerDisplayName, lastUsed: dateKey };
}
stats.byEndpoint[epKey].requests += epData.requests || 0;
stats.byEndpoint[epKey].promptTokens += epData.promptTokens || 0;
stats.byEndpoint[epKey].completionTokens += epData.completionTokens || 0;
stats.byEndpoint[epKey].cost += epData.cost || 0;
if (dateKey > (stats.byEndpoint[epKey].lastUsed || "")) stats.byEndpoint[epKey].lastUsed = dateKey;
}
}
} else {
// 24h: use live history (original logic)
const cutoff = Date.now() - PERIOD_MS["24h"];
const filtered = history.filter((e) => new Date(e.timestamp).getTime() >= cutoff);
// Handle requests with API key
if (entry.apiKey && typeof entry.apiKey === "string") {
const keyInfo = apiKeyMap[entry.apiKey];
const keyName = keyInfo?.name || entry.apiKey.slice(0, 8) + "...";
// Use full API key to avoid collisions (keys with same prefix)
const apiKeyKey = entry.apiKey;
// Group by API Key + Model + Provider combination to track different models used with the same key
const apiKeyModelKey = `${apiKeyKey}|${entry.model}|${entry.provider || 'unknown'}`;
for (const entry of filtered) {
const promptTokens = entry.tokens?.prompt_tokens || 0;
const completionTokens = entry.tokens?.completion_tokens || 0;
const entryCost = entry.cost || 0;
const providerDisplayName = providerNodeNameMap[entry.provider] || entry.provider;
if (!stats.byApiKey[apiKeyModelKey]) {
stats.byApiKey[apiKeyModelKey] = {
requests: 0,
promptTokens: 0,
completionTokens: 0,
cost: 0,
rawModel: entry.model,
provider: providerDisplayName,
apiKey: entry.apiKey,
keyName: keyName,
apiKeyKey: apiKeyKey,
lastUsed: entry.timestamp
};
stats.totalPromptTokens += promptTokens;
stats.totalCompletionTokens += completionTokens;
stats.totalCost += entryCost;
// byProvider
if (!stats.byProvider[entry.provider]) stats.byProvider[entry.provider] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0 };
stats.byProvider[entry.provider].requests++;
stats.byProvider[entry.provider].promptTokens += promptTokens;
stats.byProvider[entry.provider].completionTokens += completionTokens;
stats.byProvider[entry.provider].cost += entryCost;
// byModel
const modelKey = entry.provider ? `${entry.model} (${entry.provider})` : entry.model;
if (!stats.byModel[modelKey]) {
stats.byModel[modelKey] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0, rawModel: entry.model, provider: providerDisplayName, lastUsed: entry.timestamp };
}
const apiKeyEntry = stats.byApiKey[apiKeyModelKey];
apiKeyEntry.requests++;
apiKeyEntry.promptTokens += promptTokens;
apiKeyEntry.completionTokens += completionTokens;
apiKeyEntry.cost += entryCost;
if (new Date(entry.timestamp) > new Date(apiKeyEntry.lastUsed)) {
apiKeyEntry.lastUsed = entry.timestamp;
}
} else {
const apiKeyKey = "local-no-key";
const keyName = "Local (No API Key)";
stats.byModel[modelKey].requests++;
stats.byModel[modelKey].promptTokens += promptTokens;
stats.byModel[modelKey].completionTokens += completionTokens;
stats.byModel[modelKey].cost += entryCost;
if (new Date(entry.timestamp) > new Date(stats.byModel[modelKey].lastUsed)) stats.byModel[modelKey].lastUsed = entry.timestamp;
if (!stats.byApiKey[apiKeyKey]) {
stats.byApiKey[apiKeyKey] = {
requests: 0,
promptTokens: 0,
completionTokens: 0,
cost: 0,
rawModel: entry.model,
provider: providerDisplayName,
apiKey: null,
keyName: keyName,
apiKeyKey: apiKeyKey,
lastUsed: entry.timestamp
};
// byAccount
if (entry.connectionId) {
const accountName = connectionMap[entry.connectionId] || `Account ${entry.connectionId.slice(0, 8)}...`;
const accountKey = `${entry.model} (${entry.provider} - ${accountName})`;
if (!stats.byAccount[accountKey]) {
stats.byAccount[accountKey] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0, rawModel: entry.model, provider: providerDisplayName, connectionId: entry.connectionId, accountName, lastUsed: entry.timestamp };
}
stats.byAccount[accountKey].requests++;
stats.byAccount[accountKey].promptTokens += promptTokens;
stats.byAccount[accountKey].completionTokens += completionTokens;
stats.byAccount[accountKey].cost += entryCost;
if (new Date(entry.timestamp) > new Date(stats.byAccount[accountKey].lastUsed)) stats.byAccount[accountKey].lastUsed = entry.timestamp;
}
const apiKeyEntry = stats.byApiKey[apiKeyKey];
apiKeyEntry.requests++;
apiKeyEntry.promptTokens += promptTokens;
apiKeyEntry.completionTokens += completionTokens;
apiKeyEntry.cost += entryCost;
if (new Date(entry.timestamp) > new Date(apiKeyEntry.lastUsed)) {
apiKeyEntry.lastUsed = entry.timestamp;
// byApiKey
if (entry.apiKey && typeof entry.apiKey === "string") {
const keyInfo = apiKeyMap[entry.apiKey];
const keyName = keyInfo?.name || entry.apiKey.slice(0, 8) + "...";
const apiKeyModelKey = `${entry.apiKey}|${entry.model}|${entry.provider || "unknown"}`;
if (!stats.byApiKey[apiKeyModelKey]) {
stats.byApiKey[apiKeyModelKey] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0, rawModel: entry.model, provider: providerDisplayName, apiKey: entry.apiKey, keyName, apiKeyKey: entry.apiKey, lastUsed: entry.timestamp };
}
const ake = stats.byApiKey[apiKeyModelKey];
ake.requests++; ake.promptTokens += promptTokens; ake.completionTokens += completionTokens; ake.cost += entryCost;
if (new Date(entry.timestamp) > new Date(ake.lastUsed)) ake.lastUsed = entry.timestamp;
} else {
if (!stats.byApiKey["local-no-key"]) {
stats.byApiKey["local-no-key"] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0, rawModel: entry.model, provider: providerDisplayName, apiKey: null, keyName: "Local (No API Key)", apiKeyKey: "local-no-key", lastUsed: entry.timestamp };
}
const ake = stats.byApiKey["local-no-key"];
ake.requests++; ake.promptTokens += promptTokens; ake.completionTokens += completionTokens; ake.cost += entryCost;
if (new Date(entry.timestamp) > new Date(ake.lastUsed)) ake.lastUsed = entry.timestamp;
}
}
// By Endpoint (endpoint + model + provider combination)
const endpoint = entry.endpoint || "Unknown";
const endpointModelKey = `${endpoint}|${entry.model}|${entry.provider || 'unknown'}`;
if (!stats.byEndpoint[endpointModelKey]) {
stats.byEndpoint[endpointModelKey] = {
requests: 0,
promptTokens: 0,
completionTokens: 0,
cost: 0,
endpoint: endpoint,
rawModel: entry.model,
provider: providerDisplayName,
lastUsed: entry.timestamp
};
}
const endpointEntry = stats.byEndpoint[endpointModelKey];
endpointEntry.requests++;
endpointEntry.promptTokens += promptTokens;
endpointEntry.completionTokens += completionTokens;
endpointEntry.cost += entryCost;
if (new Date(entry.timestamp) > new Date(endpointEntry.lastUsed)) {
endpointEntry.lastUsed = entry.timestamp;
// byEndpoint
const endpoint = entry.endpoint || "Unknown";
const endpointModelKey = `${endpoint}|${entry.model}|${entry.provider || "unknown"}`;
if (!stats.byEndpoint[endpointModelKey]) {
stats.byEndpoint[endpointModelKey] = { requests: 0, promptTokens: 0, completionTokens: 0, cost: 0, endpoint, rawModel: entry.model, provider: providerDisplayName, lastUsed: entry.timestamp };
}
const epe = stats.byEndpoint[endpointModelKey];
epe.requests++; epe.promptTokens += promptTokens; epe.completionTokens += completionTokens; epe.cost += entryCost;
if (new Date(entry.timestamp) > new Date(epe.lastUsed)) epe.lastUsed = entry.timestamp;
}
}
@ -798,45 +836,48 @@ export async function getUsageStats(period = "all") {
export async function getChartData(period = "7d") {
const db = await getUsageDb();
const history = db.data.history || [];
const dailySummary = db.data.dailySummary || {};
const now = Date.now();
let bucketCount, bucketMs, labelFn;
// 24h: bucket by hour from live history
if (period === "24h") {
bucketCount = 24;
bucketMs = 3600000; // 1 hour
labelFn = (ts) => new Date(ts).toLocaleTimeString("en-US", { hour: "2-digit", minute: "2-digit", hour12: false });
} else if (period === "7d") {
bucketCount = 7;
bucketMs = 86400000;
labelFn = (ts) => new Date(ts).toLocaleDateString("en-US", { month: "short", day: "numeric" });
} else if (period === "30d") {
bucketCount = 30;
bucketMs = 86400000;
labelFn = (ts) => new Date(ts).toLocaleDateString("en-US", { month: "short", day: "numeric" });
} else {
bucketCount = 60;
bucketMs = 86400000;
labelFn = (ts) => new Date(ts).toLocaleDateString("en-US", { month: "short", day: "numeric" });
const bucketCount = 24;
const bucketMs = 3600000;
const labelFn = (ts) => new Date(ts).toLocaleTimeString("en-US", { hour: "2-digit", minute: "2-digit", hour12: false });
const startTime = now - bucketCount * bucketMs;
const buckets = Array.from({ length: bucketCount }, (_, i) => {
const ts = startTime + i * bucketMs;
return { label: labelFn(ts), tokens: 0, cost: 0 };
});
for (const entry of history) {
const entryTime = new Date(entry.timestamp).getTime();
if (entryTime < startTime || entryTime > now) continue;
const idx = Math.min(Math.floor((entryTime - startTime) / bucketMs), bucketCount - 1);
buckets[idx].tokens += (entry.tokens?.prompt_tokens || 0) + (entry.tokens?.completion_tokens || 0);
buckets[idx].cost += entry.cost || 0;
}
return buckets;
}
const startTime = now - bucketCount * bucketMs;
// 7d/30d/60d: bucket by day from dailySummary (local dates)
const bucketCount = period === "7d" ? 7 : period === "30d" ? 30 : 60;
const today = new Date();
const labelFn = (d) => d.toLocaleDateString("en-US", { month: "short", day: "numeric" });
const buckets = Array.from({ length: bucketCount }, (_, i) => {
const ts = startTime + i * bucketMs;
return { label: labelFn(ts), tokens: 0, cost: 0, _ts: ts };
const d = new Date(today);
d.setDate(d.getDate() - (bucketCount - 1 - i));
const dateKey = `${d.getFullYear()}-${String(d.getMonth() + 1).padStart(2, "0")}-${String(d.getDate()).padStart(2, "0")}`;
const dayData = dailySummary[dateKey];
return {
label: labelFn(d),
tokens: dayData ? (dayData.promptTokens || 0) + (dayData.completionTokens || 0) : 0,
cost: dayData ? (dayData.cost || 0) : 0,
};
});
for (const entry of history) {
const entryTime = new Date(entry.timestamp).getTime();
if (entryTime < startTime || entryTime > now) continue;
const idx = Math.min(Math.floor((entryTime - startTime) / bucketMs), bucketCount - 1);
const promptTokens = entry.tokens?.prompt_tokens || 0;
const completionTokens = entry.tokens?.completion_tokens || 0;
buckets[idx].tokens += promptTokens + completionTokens;
// Use pre-stored cost if available, else 0
buckets[idx].cost += entry.cost || 0;
}
return buckets.map(({ label, tokens, cost }) => ({ label, tokens, cost }));
return buckets;
}
// Re-export request details functions from new SQLite-based module

View file

@ -10,6 +10,7 @@ import { MEDIA_PROVIDER_KINDS } from "@/shared/constants/providers";
import Button from "./Button";
import { ConfirmModal } from "./Modal";
// const VISIBLE_MEDIA_KINDS = ["embedding", "image", "imageToText", "tts", "stt", "webSearch", "webFetch", "video", "music"];
const VISIBLE_MEDIA_KINDS = ["embedding", "tts"];
const navItems = [

View file

@ -9,16 +9,16 @@ export const FREE_PROVIDERS = {
// codebuddy: { id: "codebuddy", alias: "cb", name: "CodeBuddy", icon: "smart_toy", color: "#006EFF" },
// qoder: { id: "qoder", alias: "qd", name: "Qoder AI", icon: "water_drop", color: "#EC4899" },
iflow: { id: "iflow", alias: "if", name: "iFlow AI", icon: "water_drop", color: "#6366F1" },
opencode: { id: "opencode", alias: "oc", name: "OpenCode", icon: "terminal", color: "#E87040", textIcon: "OC", noAuth: true },
opencode: { id: "opencode", alias: "oc", name: "OpenCode", icon: "terminal", color: "#E87040", textIcon: "OC", noAuth: true, passthroughModels: true, modelsFetcher: { url: "https://opencode.ai/zen/v1/models", type: "opencode-free" } },
};
// Free Tier Providers (has free access but may require account/API key)
export const FREE_TIER_PROVIDERS = {
openrouter: { id: "openrouter", alias: "openrouter", name: "OpenRouter", icon: "router", color: "#F97316", textIcon: "OR", website: "https://openrouter.ai", notice: { text: "Free tier: 27+ free models, no credit card needed, 200 req/day. After $10 credit: 1,000 req/day.", apiKeyUrl: "https://openrouter.ai/settings/keys" }, modelsFetcher: { url: "https://openrouter.ai/api/v1/models", type: "openrouter-free" }, passthroughModels: true, serviceKinds: ["llm", "embedding"] },
openrouter: { id: "openrouter", alias: "openrouter", name: "OpenRouter", icon: "router", color: "#F97316", textIcon: "OR", website: "https://openrouter.ai", notice: { text: "Free tier: 27+ free models, no credit card needed, 200 req/day. After $10 credit: 1,000 req/day.", apiKeyUrl: "https://openrouter.ai/settings/keys" }, modelsFetcher: { url: "https://openrouter.ai/api/v1/models", type: "openrouter-free" }, passthroughModels: true, serviceKinds: ["llm", "embedding", "tts", "imageToText"] },
nvidia: { id: "nvidia", alias: "nvidia", name: "NVIDIA NIM", icon: "developer_board", color: "#76B900", textIcon: "NV", website: "https://developer.nvidia.com/nim", notice: { text: "Free access for NVIDIA Developer Program members (prototyping & testing).", apiKeyUrl: "https://build.nvidia.com/settings/api-keys" } },
ollama: { id: "ollama", alias: "ollama", name: "Ollama Cloud", icon: "cloud", color: "#ffffffff", textIcon: "OL", website: "https://ollama.com", notice: { text: "Free tier: light usage, 1 cloud model at a time (limits reset every 5h & 7d). Pro $20/mo · Max $100/mo.", apiKeyUrl: "https://ollama.com/settings/keys" } },
vertex: { id: "vertex", alias: "vx", name: "Vertex AI", icon: "cloud", color: "#4285F4", textIcon: "VX", website: "https://cloud.google.com/vertex-ai", notice: { text: "New Google Cloud accounts get $300 free credits. Requires GCP project + Service Account with Vertex AI API enabled.", apiKeyUrl: "https://console.cloud.google.com/iam-admin/serviceaccounts" } },
gemini: { id: "gemini", alias: "gemini", name: "Gemini", icon: "diamond", color: "#4285F4", textIcon: "GE", website: "https://ai.google.dev", serviceKinds: ["llm", "embedding"] },
gemini: { id: "gemini", alias: "gemini", name: "Gemini", icon: "diamond", color: "#4285F4", textIcon: "GE", website: "https://ai.google.dev", serviceKinds: ["llm", "embedding", "image", "imageToText", "webSearch"] },
};
// Thinking config definitions
@ -54,20 +54,20 @@ export const OAUTH_PROVIDERS = {
export const APIKEY_PROVIDERS = {
glm: { id: "glm", alias: "glm", name: "GLM Coding", icon: "code", color: "#2563EB", textIcon: "GL", website: "https://open.bigmodel.cn" },
"glm-cn": { id: "glm-cn", alias: "glm-cn", name: "GLM (China)", icon: "code", color: "#DC2626", textIcon: "GC", website: "https://open.bigmodel.cn" },
kimi: { id: "kimi", alias: "kimi", name: "Kimi", icon: "psychology", color: "#1E3A8A", textIcon: "KM", website: "https://kimi.moonshot.cn" },
minimax: { id: "minimax", alias: "minimax", name: "Minimax Coding", icon: "memory", color: "#7C3AED", textIcon: "MM", website: "https://www.minimaxi.com" },
kimi: { id: "kimi", alias: "kimi", name: "Kimi", icon: "psychology", color: "#1E3A8A", textIcon: "KM", website: "https://kimi.moonshot.cn", serviceKinds: ["llm", "webSearch"] },
minimax: { id: "minimax", alias: "minimax", name: "Minimax Coding", icon: "memory", color: "#7C3AED", textIcon: "MM", website: "https://www.minimaxi.com", serviceKinds: ["llm", "image", "imageToText", "webSearch"] },
"minimax-cn": { id: "minimax-cn", alias: "minimax-cn", name: "Minimax (China)", icon: "memory", color: "#DC2626", textIcon: "MC", website: "https://www.minimaxi.com" },
alicode: { id: "alicode", alias: "alicode", name: "Alibaba", icon: "cloud", color: "#FF6A00", textIcon: "ALi" },
"alicode-intl": { id: "alicode-intl", alias: "alicode-intl", name: "Alibaba Intl", icon: "cloud", color: "#FF6A00", textIcon: "ALi" },
openai: { id: "openai", alias: "openai", name: "OpenAI", icon: "auto_awesome", color: "#10A37F", textIcon: "OA", website: "https://platform.openai.com", serviceKinds: ["llm", "embedding", "tts"], thinkingConfig: THINKING_CONFIG.effort },
anthropic: { id: "anthropic", alias: "anthropic", name: "Anthropic", icon: "smart_toy", color: "#D97757", textIcon: "AN", website: "https://console.anthropic.com", serviceKinds: ["llm"] },
openai: { id: "openai", alias: "openai", name: "OpenAI", icon: "auto_awesome", color: "#10A37F", textIcon: "OA", website: "https://platform.openai.com", serviceKinds: ["llm", "embedding", "tts", "image", "imageToText", "webSearch"], thinkingConfig: THINKING_CONFIG.effort },
anthropic: { id: "anthropic", alias: "anthropic", name: "Anthropic", icon: "smart_toy", color: "#D97757", textIcon: "AN", website: "https://console.anthropic.com", serviceKinds: ["llm", "imageToText"] },
deepseek: { id: "deepseek", alias: "ds", name: "DeepSeek", icon: "bolt", color: "#4D6BFE", textIcon: "DS", website: "https://deepseek.com" },
groq: { id: "groq", alias: "groq", name: "Groq", icon: "speed", color: "#F55036", textIcon: "GQ", website: "https://groq.com" },
xai: { id: "xai", alias: "xai", name: "xAI (Grok)", icon: "auto_awesome", color: "#1DA1F2", textIcon: "XA", website: "https://x.ai" },
mistral: { id: "mistral", alias: "mistral", name: "Mistral", icon: "air", color: "#FF7000", textIcon: "MI", website: "https://mistral.ai" },
perplexity: { id: "perplexity", alias: "pplx", name: "Perplexity", icon: "search", color: "#20808D", textIcon: "PP", website: "https://www.perplexity.ai" },
groq: { id: "groq", alias: "groq", name: "Groq", icon: "speed", color: "#F55036", textIcon: "GQ", website: "https://groq.com", serviceKinds: ["llm", "imageToText"] },
xai: { id: "xai", alias: "xai", name: "xAI (Grok)", icon: "auto_awesome", color: "#1DA1F2", textIcon: "XA", website: "https://x.ai", serviceKinds: ["llm", "imageToText", "webSearch"] },
mistral: { id: "mistral", alias: "mistral", name: "Mistral", icon: "air", color: "#FF7000", textIcon: "MI", website: "https://mistral.ai", serviceKinds: ["llm", "imageToText"] },
perplexity: { id: "perplexity", alias: "pplx", name: "Perplexity", icon: "search", color: "#20808D", textIcon: "PP", website: "https://www.perplexity.ai", serviceKinds: ["llm", "webSearch"] },
together: { id: "together", alias: "together", name: "Together AI", icon: "group_work", color: "#0F6FFF", textIcon: "TG", website: "https://www.together.ai" },
fireworks: { id: "fireworks", alias: "fireworks", name: "Fireworks AI", icon: "local_fire_department", color: "#7B2EF2", textIcon: "FW", website: "https://fireworks.ai" },
cerebras: { id: "cerebras", alias: "cerebras", name: "Cerebras", icon: "memory", color: "#FF4F00", textIcon: "CB", website: "https://www.cerebras.ai" },
@ -75,7 +75,7 @@ export const APIKEY_PROVIDERS = {
nebius: { id: "nebius", alias: "nebius", name: "Nebius AI", icon: "cloud", color: "#6C5CE7", textIcon: "NB", website: "https://nebius.com" },
siliconflow: { id: "siliconflow", alias: "siliconflow", name: "SiliconFlow", icon: "cloud_queue", color: "#5B6EF5", textIcon: "SF", website: "https://cloud.siliconflow.com" },
hyperbolic: { id: "hyperbolic", alias: "hyp", name: "Hyperbolic", icon: "bolt", color: "#00D4FF", textIcon: "HY", website: "https://hyperbolic.xyz" },
deepgram: { id: "deepgram", alias: "dg", name: "Deepgram", icon: "mic", color: "#13EF93", textIcon: "DG", website: "https://deepgram.com", serviceKinds: ["stt"] },
deepgram: { id: "deepgram", alias: "dg", name: "Deepgram", icon: "mic", color: "#13EF93", textIcon: "DG", website: "https://deepgram.com", serviceKinds: ["stt", "imageToText"] },
assemblyai: { id: "assemblyai", alias: "aai", name: "AssemblyAI", icon: "record_voice_over", color: "#0062FF", textIcon: "AA", website: "https://assemblyai.com", serviceKinds: ["stt"] },
nanobanana: { id: "nanobanana", alias: "nb", name: "NanoBanana", icon: "image", color: "#FFD700", textIcon: "NB", website: "https://nanobananaapi.ai", serviceKinds: ["image"] },
elevenlabs: { id: "elevenlabs", alias: "el", name: "ElevenLabs", icon: "record_voice_over", color: "#6C47FF", textIcon: "EL", website: "https://elevenlabs.io", serviceKinds: ["tts"] },
@ -86,20 +86,29 @@ export const APIKEY_PROVIDERS = {
"edge-tts": { id: "edge-tts", alias: "edge-tts", name: "Edge TTS", icon: "record_voice_over", color: "#0078D4", textIcon: "ET", serviceKinds: ["tts"], noAuth: true },
sdwebui: { id: "sdwebui", alias: "sdwebui", name: "SD WebUI", icon: "brush", color: "#FF7043", textIcon: "SD", website: "https://github.com/AUTOMATIC1111/stable-diffusion-webui", serviceKinds: ["image"] },
comfyui: { id: "comfyui", alias: "comfyui", name: "ComfyUI", icon: "account_tree", color: "#4CAF50", textIcon: "CF", website: "https://github.com/comfyanonymous/ComfyUI", serviceKinds: ["image"] },
huggingface: { id: "huggingface", alias: "hf", name: "HuggingFace", icon: "face", color: "#FFD21E", textIcon: "HF", website: "https://huggingface.co", serviceKinds: ["image", "tts"], hiddenKinds: ["tts"] },
huggingface: { id: "huggingface", alias: "hf", name: "HuggingFace", icon: "face", color: "#FFD21E", textIcon: "HF", website: "https://huggingface.co", serviceKinds: ["image", "imageToText", "tts"], hiddenKinds: ["tts"] },
chutes: { id: "chutes", alias: "ch", name: "Chutes AI", icon: "water_drop", color: "#ffffffff", textIcon: "CH", website: "https://chutes.ai" },
"ollama-local": { id: "ollama-local", alias: "ollama-local", name: "Ollama Local", icon: "cloud", color: "#ffffffff", textIcon: "OL", website: "https://ollama.com" },
"vertex-partner": { id: "vertex-partner", alias: "vxp", name: "Vertex Partner", icon: "cloud", color: "#34A853", textIcon: "VP", website: "https://cloud.google.com/vertex-ai/generative-ai/docs/partner-models/use-partner-models" },
tavily: { id: "tavily", alias: "tavily", name: "Tavily", icon: "search", color: "#5B21B6", textIcon: "TV", website: "https://tavily.com", serviceKinds: ["webSearch"] },
"brave-search": { id: "brave-search", alias: "brave", name: "Brave Search", icon: "travel_explore", color: "#FB542B", textIcon: "BR", website: "https://brave.com/search/api", serviceKinds: ["webSearch"] },
serper: { id: "serper", alias: "serper", name: "Serper", icon: "search", color: "#4F46E5", textIcon: "SP", website: "https://serper.dev", serviceKinds: ["webSearch"] },
exa: { id: "exa", alias: "exa", name: "Exa", icon: "manage_search", color: "#2563EB", textIcon: "EX", website: "https://exa.ai", serviceKinds: ["webSearch"] },
searxng: { id: "searxng", alias: "searxng", name: "SearXNG", icon: "saved_search", color: "#3B82F6", textIcon: "SX", website: "https://docs.searxng.org", serviceKinds: ["webSearch"], noAuth: true },
firecrawl: { id: "firecrawl", alias: "firecrawl", name: "Firecrawl", icon: "local_fire_department", color: "#F59E0B", textIcon: "FC", website: "https://firecrawl.dev", serviceKinds: ["webFetch"] },
};
// Media provider kinds — each kind maps to a route and endpoint config
export const MEDIA_PROVIDER_KINDS = [
{ id: "embedding", label: "Embedding", icon: "data_array", endpoint: { method: "POST", path: "/v1/embeddings" } },
{ id: "image", label: "Image", icon: "image", endpoint: { method: "POST", path: "/v1/images/generations" } },
{ id: "tts", label: "Text To Speech", icon: "record_voice_over", endpoint: { method: "POST", path: "/v1/audio/speech" } },
{ id: "stt", label: "STT", icon: "mic", endpoint: { method: "POST", path: "/v1/audio/transcriptions" } },
{ id: "video", label: "Video", icon: "movie", endpoint: { method: "POST", path: "/v1/video/generations" } },
{ id: "music", label: "Music", icon: "music_note", endpoint: { method: "POST", path: "/v1/audio/music" } },
{ id: "embedding", label: "Embedding", icon: "data_array", endpoint: { method: "POST", path: "/v1/embeddings" } },
{ id: "image", label: "Text to Image", icon: "brush", endpoint: { method: "POST", path: "/v1/images/generations" } },
{ id: "imageToText", label: "Image to Text", icon: "image_search", endpoint: { method: "POST", path: "/v1/images/understanding" } },
{ id: "tts", label: "Text To Speech", icon: "record_voice_over", endpoint: { method: "POST", path: "/v1/audio/speech" } },
{ id: "stt", label: "STT", icon: "mic", endpoint: { method: "POST", path: "/v1/audio/transcriptions" } },
{ id: "webSearch", label: "Web Search", icon: "travel_explore", endpoint: { method: "POST", path: "/v1/search" } },
{ id: "webFetch", label: "Web Fetch", icon: "language", endpoint: { method: "POST", path: "/v1/web/fetch" } },
{ id: "video", label: "Video", icon: "movie", endpoint: { method: "POST", path: "/v1/video/generations" } },
{ id: "music", label: "Music", icon: "music_note", endpoint: { method: "POST", path: "/v1/audio/music" } },
];
export const OPENAI_COMPATIBLE_PREFIX = "openai-compatible-";

View file

@ -13,9 +13,19 @@ export const TTS_PROVIDER_CONFIG = {
hasLanguageDropdown: false,
hasModelSelector: true,
hasBrowseButton: false,
voiceSource: "hardcoded", // from providerModels
voiceSource: "hardcoded",
modelKey: "openai-tts-models",
voiceKey: "openai-tts-voices",
voicesPerModel: true,
},
"openrouter": {
hasLanguageDropdown: false,
hasModelSelector: true,
hasBrowseButton: false,
voiceSource: "hardcoded",
modelKey: "openrouter-tts-models",
voiceKey: "openrouter-tts-voices",
voicesPerModel: true,
},
"elevenlabs": {
hasLanguageDropdown: false,

View file

@ -1,28 +1,14 @@
// Fetch and cache suggested models for providers that expose a public models API
// Designed to be extensible: add new types in FILTERS below
// Fetches via backend proxy to avoid CORS issues
const CACHE_TTL_MS = 10 * 60 * 1000; // 10 minutes
const cache = new Map(); // key: fetcher.url → { data, expiresAt }
const FILTERS = {
// Free models with context >= 200k tokens
"openrouter-free": (models) =>
models
.filter(
(m) =>
m.pricing?.prompt === "0" &&
m.pricing?.completion === "0" &&
m.context_length >= 200000
)
.map((m) => ({ id: m.id, name: m.name, contextLength: m.context_length }))
.sort((a, b) => b.contextLength - a.contextLength),
};
/**
* Fetch suggested models for a provider using its modelsFetcher config.
* Results are cached in-memory for CACHE_TTL_MS.
* @param {{ url: string, type: string }} fetcher
* @returns {Promise<Array<{ id: string, name: string, contextLength: number }>>}
* @returns {Promise<Array<{ id: string, name: string, contextLength?: number }>>}
*/
export async function fetchSuggestedModels(fetcher) {
if (!fetcher?.url || !fetcher?.type) return [];
@ -31,12 +17,11 @@ export async function fetchSuggestedModels(fetcher) {
if (cached && Date.now() < cached.expiresAt) return cached.data;
try {
const res = await fetch(fetcher.url);
const params = new URLSearchParams({ url: fetcher.url, type: fetcher.type });
const res = await fetch(`/api/providers/suggested-models?${params}`);
if (!res.ok) return [];
const json = await res.json();
const raw = json.data ?? json.models ?? json;
const filter = FILTERS[fetcher.type];
const data = filter ? filter(Array.isArray(raw) ? raw : []) : [];
const data = json.data ?? [];
cache.set(fetcher.url, { data, expiresAt: Date.now() + CACHE_TTL_MS });
return data;
} catch {

View file

@ -10,7 +10,7 @@ import { HTTP_STATUS } from "open-sse/config/runtimeConfig.js";
import * as log from "../utils/logger.js";
// Providers that require stored credentials (not noAuth)
const CREDENTIALED_PROVIDERS = new Set(["openai", "elevenlabs"]);
const CREDENTIALED_PROVIDERS = new Set(["openai", "elevenlabs", "openrouter"]);
export async function handleTts(request) {
let body;