- Cap maximum cooldown for rate limit handling in account unavailability and single-model chat flows

- Dynamic custom model fetching for model selection

src/sse/handlers/chat.js

@@ -206,6 +206,7 @@ async function handleSingleModelChat(body, modelStr, clientRawRequest = null, re
       userAgent,
       apiKey,
       ccFilterNaming: !!chatSettings.ccFilterNaming,
+      rtkEnabled: !!chatSettings.rtkEnabled,
       providerThinking,
       // Detect source format by endpoint + body
       sourceFormatOverride: request?.url ? detectFormatByEndpoint(new URL(request.url).pathname, body) : null,

src/sse/services/auth.js

@@ -1,6 +1,7 @@
 import { getProviderConnections, validateApiKey, updateProviderConnection, getSettings } from "@/lib/localDb";
 import { resolveConnectionProxyConfig } from "@/lib/network/connectionProxy";
 import { formatRetryAfter, checkFallbackError, isModelLockActive, buildModelLockUpdate, getEarliestModelLockUntil } from "open-sse/services/accountFallback.js";
+import { MAX_RATE_LIMIT_COOLDOWN_MS } from "open-sse/config/errorConfig.js";
 import { resolveProviderId, FREE_PROVIDERS } from "@/shared/constants/providers.js";
 import * as log from "../utils/logger.js";
 
@@ -179,7 +180,7 @@ export async function markAccountUnavailable(connectionId, status, errorText, pr
   let shouldFallback, cooldownMs, newBackoffLevel;
   if (resetsAtMs && resetsAtMs > Date.now()) {
     shouldFallback = true;
-    cooldownMs = resetsAtMs - Date.now();
+    cooldownMs = Math.min(resetsAtMs - Date.now(), MAX_RATE_LIMIT_COOLDOWN_MS);
     newBackoffLevel = 0;
   } else {
     ({ shouldFallback, cooldownMs, newBackoffLevel } = checkFallbackError(status, errorText, backoffLevel));