Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

Codex local model output limit fixed via model catalog

AI summary

A Juejin AI tutorial reports that Codex CLI 0.169.0 stopped during a long script edit after hitting a 3,208-token single-output limit, despite the local Qwen model having a 131,072-token context window. The author’s workaround adds a model-catalog.json with truncation_policy.limit set to 16,384, points config.toml to it, and validates the change with longer output tests; the results come from one Windows 11 setup.

In-depth insight

The tutorial provides a concrete configuration path and acceptance checks for Codex users whose local models are limited by fallback metadata rather than server capacity.

我把 codex 接到本机 Strata 上的 Qwen3.8 模型(RTX 5060 Ti 16GB + 64GB 内存,131072 上下文,服务在 127.0.0.1:8080)。

日常问答正常,改小文件也正常。但让它改一个两百多行的脚本时,它半途停下,屏幕上留两行:

writing a tool call: 2368 of max 3208 tokens, 61.6 tok/s
Incomplete response returned, reason: max_output_tokens

上下文明明 131072,单次输出却只给 3208,占 2.4%。

codex 输出被 max_output_tokens 截断,命令行显示 2368 of max 3208 tokens

这篇教程记录把输出长度增加到 16384 的完整过程。你照着做就行:先判断自己的情况,再用脚本一键改,最后按三个判据验收。如果脚本跑不通,后面还有手动路径和一张报错对照表。

#Codex CLI 0.169.0#Strata 0.1.61#Qwen3.8-Flash-Next IQ3_S#Claude Code#RTX 5060 Ti

Original · 11 min readRead the original
Group chat

No comments yet. Start the conversation.