{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreigm4gwvgypeggqwgs3uycawasauj2zdr4x2276tgb4wwmy2r63w3m",
    "uri": "at://did:plc:lk3jfj3zq4k4wxnk474axylu/app.bsky.feed.post/3miqkyg3tgfl2"
  },
  "path": "/t/feature-request-let-codex-build-and-embed-small-task-specific-local-models/1378578#post_1",
  "publishedAt": "2026-04-05T09:57:25.000Z",
  "site": "https://community.openai.com",
  "textContent": "Hello OpenAI team,\n\nI would like to request a Codex feature for local-first software development.\n\nFeature idea:\nAllow Codex to optionally prepare, adapt, quantize, and integrate a small task-specific local model into the applications it builds.\n\nExample:\nIf I ask Codex to build a debt management desktop app, it should be able to optionally prepare a lightweight local model for that app’s specific use case and integrate it into the generated project so core AI features can work offline.\n\nWhy this would be valuable:\n\n  * offline and privacy-friendly app experiences\n  * less dependency on external APIs for end users\n  * more useful generated apps\n  * stronger local-first developer workflows\n  * better support for specialized software\n\n\n\nImportant note:\nThis could be controlled by plan limits or compute limits, so higher tiers could support larger local model workflows while standard paid tiers support smaller task-specific models.\n\nWhat I’m asking for:\n\n  * optional local model preparation inside Codex\n  * optional task adaptation / lightweight fine-tuning flow\n  * optional quantization/export flow\n  * ability to integrate the resulting lightweight model into generated desktop or web apps\n  * clear controls for model size, runtime limits, and deployment target\n\n\n\nThis would make Codex much more powerful for developers building real local-first software.\n\nThanks.",
  "title": "Feature Request: Let Codex Build and Embed Small Task-Specific Local Models"
}