{"slug": "claude-code-vs-codex-i-get-about-150-vs-30-tokens-sec", "title": "Claude Code vs Codex: I Get About 150 vs 30 Tokens/sec", "summary": "A power user reports averaging roughly 150 tokens per second with Claude Code versus about 30 tokens per second with OpenAI's Codex on flagship models, and recommends Claude Code over Codex for almost everybody on that basis. The author, who says he typically runs up to 15 agents in parallel, notes 15 Codex agents at 30 tokens per second aggregate to 450 tokens per second, matching three Claude Code agents at 150 tokens per second, and cautions the figures are self-reported rather than a controlled benchmark.", "body_md": "# Claude Code vs Codex: I Get About 150 vs 30 Tokens/sec\n\nI strongly recommend Claude Code over Codex for almost everybody. The big issue for me is speed.\n\nWith Codex, I average around 30 tokens per second on the flagship models I use. With Claude Code, I average around 150 tokens per second.\n\nWhat does this mean in practice? Well, as a power user who’s usually running up to 15 agents in parallel, 15 Codex agents at 30 tokens per second add up to 450 tokens per second. Three Claude Code agents at 150 tokens per second add up to the same total.\n\nThose are my reported rates, not a controlled benchmark. The arithmetic is about aggregate token generation, not equal code quality or five times as many completed tasks. Models and settings change rates, and faster modes can use more of your allowance or cost more: [https://developers.openai.com/codex/speed/](https://developers.openai.com/codex/speed/) \n\n## Running 15 agents in parallel is demanding\n\nPeople think, oh, the agent is doing the work. It doesn’t matter how long it takes. But the reality is, it does. Running 15 agents in parallel is demanding. It’s not easy work. My productivity takes a significant hit when I’m coordinating that many agents.\n\nWhen the token-rate arithmetic is 15 agents with Codex versus three with Claude Code, you know which one I’m going to prefer. Benchmark scores and project results don’t show how demanding that coordination feels.\n\nYou just feel like, oh shit, Claude Code just gets shit done. For me, it’s not about the system prompt. It’s not about the skills. It’s not about the tools. It’s the speed.", "url": "https://wpnews.pro/news/claude-code-vs-codex-i-get-about-150-vs-30-tokens-sec", "canonical_source": "https://ziva.sh/blogs/claude-code-vs-codex", "published_at": "2026-10-07 00:00:00+00:00", "updated_at": "2026-10-07 12:18:05.951538+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "large-language-models", "developer-tools"], "entities": ["Claude Code", "Codex", "OpenAI", "Anthropic"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/claude-code-vs-codex-i-get-about-150-vs-30-tokens-sec", "markdown": "https://wpnews.pro/news/claude-code-vs-codex-i-get-about-150-vs-30-tokens-sec.md", "text": "https://wpnews.pro/news/claude-code-vs-codex-i-get-about-150-vs-30-tokens-sec.txt", "jsonld": "https://wpnews.pro/news/claude-code-vs-codex-i-get-about-150-vs-30-tokens-sec.jsonld"}}