{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreider5nnoxxautl6jjbvfpfftnxfe3bc4czj76mkmptvgprgu7q75u",
"uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3mospxl7hdk42"
},
"path": "/t/how-to-use-multiple-models-in-a-codex-workflow/1384269#post_2",
"publishedAt": "2026-06-21T14:35:06.000Z",
"site": "https://community.openai.com",
"tags": [
"@report.md"
],
"textContent": "You can change the model between turns using /model.\n\nThere are two ways to approach this:\n\n 1. Start with gpt-5.4-mini, then switch to gpt-5.5 in the next turn. Note that this will send all previous turns in the thread to gpt-5.5 as uncached context.\n\n 2. The preferred method, if your goal is to save the most usage limits:\n\n * Use gpt-5.4-mini to create a file, such as report.md, containing the results.\n * Prompt: `Please analyze <your giant task> and save the report in report.md.`\n * Start a new thread with /new, switch to gpt-5.5, and use @report.md to refer to the file. This discards all previous reasoning and initial inputs, using only the summarized report instead, which saves the most tokens.\n * Prompt: `@report.md <your gpt-5 task>`\n\n",
"title": "How to use multiple models in a Codex workflow"
}