feat: Make MAX_TOKENS and MAX_RESPONSE_SEGMENTS configurable

Added MAX_TOKENS and MAX_RESPONSE_SEGMENTS to .env.example and documentation. Updated constants to read from environment variables, allowing configuration of token and response segment limits.

.env.example

@@ -10,6 +10,10 @@ OPERATING_ENV=production
 STORAGE_DIR=/public/uploads
 # Maximum upload size for attachments
 MAX_UPLOAD_SIZE_MB=5
+# The number of segments when reaching the maximum Tokens count
+MAX_RESPONSE_SEGMENTS=5
+# Default token usage for a single chat
+MAX_TOKENS=8000
 
 # Example Context Values for qwen2.5-coder:32b
 #

app/lib/.server/llm/constants.ts

@@ -1,5 +1,7 @@
 // see https://docs.anthropic.com/en/docs/about-claude/models
-export const MAX_TOKENS = 8000;
+export const MAX_TOKENS = process.env.MAX_TOKENS ? parseInt(process.env.MAX_TOKENS) : 8000;
 
 // limits the number of model responses that can be returned in a single request
-export const MAX_RESPONSE_SEGMENTS = 3;
+export const MAX_RESPONSE_SEGMENTS = process.env.MAX_RESPONSE_SEGMENTS
+  ? parseInt(process.env.MAX_RESPONSE_SEGMENTS)
+  : 5;

app/routes/api.chat/chat.server.ts

@@ -331,7 +331,7 @@ export async function chatAction({ request, userId }: ChatActionArgs) {
 
       const continueMessage = async (text: string) => {
         logger.info(
-          `达到最大 token 限制 (${MAX_TOKENS}): 继续消息, 还可以响应 (${MAX_RESPONSE_SEGMENTS - streamSwitches} 个分段)`,
+          `达到最大 token 限制 (${DEFAULT_MODEL_DETAILS?.maxTokenAllowed || MAX_TOKENS}): 继续消息, 还可以响应 (${MAX_RESPONSE_SEGMENTS - streamSwitches} 个分段)`,
         );
         messages.push({
           id: generateId(),

docs/content/configuration.md

@@ -24,6 +24,8 @@ UPage 使用环境变量进行配置。您可以通过以下方式设置环境
 | `USAGE_LOG_FILE` | 是否开启文件日志 | `true` | 否 |
 | `MAX_UPLOAD_SIZE_MB` | 附件上传的最大大小 (MB) | `5` | 否 |
 | `STORAGE_DIR` | 资源文件存储位置 | `/app/storage` | 否 |
+| `MAX_RESPONSE_SEGMENTS` | 最大响应分段数 | `5` | 否 |
+| `MAX_TOKENS` | 最大 token 数 | `8000` | 否 |
 
 ## AI 提供商配置