#codexcli

Originally published on andrew.ooo — visit the original for any updates, code snippets that aged...

Six lines of TOML put Qwen 3.8 Max in Codex CLI. Three real tasks cost $0.08 against $0.54 on GPT-5.5, and the 258K context cap has exactly one fix.

opencodex points Codex CLI and Claude Code at any vendor's model. Here's the verified per-token cost of each one — including the September Sonnet 5 price rise Anthropic just cancelled.

Originally published at norvik.tech Introduction Deep dive into Codex CLI's custom...

Codex CLI auth breaks in nine distinct ways and only some are literal 401s. The message body identifies the cause. All cases reproduced on Codex CLI 0.146.0.

OpenCode (model-agnostic, MIT) vs Codex CLI (OpenAI-native, Rust). Choose by your stack: multi-model freedom or turnkey safety.

One AI coding-agent harness for Claude Code, Codex and Copilot CLI: a 2026 decision...

OpenCodex is a local npm proxy that routes Codex CLI to Anthropic, Gemini, Ollama, and 40+ providers — no patches to Codex. Install and run

AGENTS.md audit: which of 13 coding tools genuinely read the spec vs. fake it, the real merge/precedence rules, and where its governance story is thinner than it looks.

Claude Code vs Codex CLI vs Gemini CLI in July 2026: verified pricing and models, sandbox internals down to seccomp, MCP support, and the closed-source Antigravity fallout.

Claude Code vs Codex CLI, compared for 2026: pricing, models, MCP and extensibility, open source, and which terminal AI coding agent you should actually pick.

Terminal-Bench 2.1 in June 2026: The #1 Model Is One You Can't Use — Here's the Leaderboard That Actually Matters