[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-75878":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":15,"forks30d":15,"starsTrendScore":19,"compositeScore":20,"rankGlobal":10,"rankLanguage":10,"license":21,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":22,"hasPages":22,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":15,"starSnapshotCount":15,"syncStatus":28,"lastSyncTime":29,"discoverSource":30},75878,"nolanx.ai","nolanx-ai\u002Fnolanx.ai","nolanx-ai","Nolanx, Open-sourced AI Netflix.","",null,"Python",1003,69,55,0,17,194,938,103,18.54,"MIT License",false,"main",[],"2026-06-12 02:03:37","# NolanX\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fpub-fb1cce7145174a7b9989934451fb797a.r2.dev\u002Fnolanx\u002Fnolanx.mp4\">\n    \u003Cimg src=\"apps\u002Fweb\u002Fpublic\u002Freadme\u002Fnolanx-preview.gif\" alt=\"NolanX demo video\" \u002F>\n  \u003C\u002Fa>\n\u003C\u002Fp>\n\nOpen-source long-runtime multi-modal agent infrastructure for filmmaking.\n\n![NolanX Home](apps\u002Fweb\u002Fpublic\u002Fnolanx-home.png)\n\n![NolanX Timeline](apps\u002Fweb\u002Fpublic\u002Fnolanx-timeline.png)\n\nNolanX is the system behind [`nolanx.ai`](https:\u002F\u002Fnolanx.ai\u002F). It is designed to push video models far beyond their native single-generation limits and keep long creative runs coherent across language, image, audio, and video.\n\nWith the right model stack, NolanX can turn short-form generators such as Seedance 2.0, Kling 03, Google Veo 3.1 Lite, LTX 2.3, PixVerse v6, Wan 2.7, and others into much longer production pipelines: 5 minutes, 30 minutes, even 1 hour.\n\nIt also keeps consistency across large agent workloads: story logic, screenplay, script breakdown, character views, scenes, props, sound, auditions, world assets, timeline editing, storyboard structure, dialogue, performance detail, and directorial reasoning.\n\n## Why this exists\n\nThere are only two of us maintaining [`reelmind.ai`](https:\u002F\u002Freelmind.ai\u002F) and [`nolanx.ai`](https:\u002F\u002Fnolanx.ai\u002F).\n\nI’m a 10-year entrepreneur, now living as an AI nomad between Serbia, Europe, and North Africa.\n\nOne day I told my partner Keii:\n\n> If you can build something where one click, just one click, generates a movie like Nezha 3 or Interstellar 2, I’ll call you Dad.\n\nEventually, I built that technology myself.\n\nI believe almost every serious AI film builder will end up extending models this way. And I believe many AI companies will too.\n\nThat is why I chose to open-source NolanX.\n\nSorry, Netflix. When I finally found the right disruptive slogan, I thought of you.\n\nHave fun.\n\n## Architecture\n\n![NolanX Runtime Architecture](apps\u002Fweb\u002Fpublic\u002Fnolanx-runtime.png)\n\nThis repo is trimmed to the NolanX path. Community, membership, admin, and credit logic are out of scope here.\n\n## Use NolanX\n\n1. Use [`nolanx.ai`](https:\u002F\u002Fnolanx.ai\u002F) directly with membership or credit packs\n   Just open [nolanx.ai](https:\u002F\u002Fnolanx.ai\u002F).\n\n2. Use the hosted API route if you want NolanX runtime behavior without wiring every provider yourself\n   - `Text`: [OpenRouter](https:\u002F\u002Fopenrouter.ai\u002Fworkspaces\u002Fdefault\u002Fkeys)\n   - `Image`: [FAL](https:\u002F\u002Ffal.ai\u002Fdashboard\u002Fkeys)\n   - `Video`: [ReelMind](https:\u002F\u002Freelmind.ai\u002Fplatform), our own API platform with a 5% markup\n   - `Optional R2`: [Cloudflare R2](https:\u002F\u002Fdash.cloudflare.com\u002F?to=\u002F:account\u002Fr2\u002Foverview)\n\n3. Fully self-host it locally or privately\n   Run the full stack yourself: web, API, and agent runtime.\n   - Configure `apps\u002Fagent\u002Fconfig.toml` first\n   - Use root `.env` only for local host\u002Fport overrides\n   - Use `\u002Fnolanx` `Runtime Keys` when you want to edit runtime overlays\n   - Replace providers, priorities, and storage paths as needed\n\n## Quick start\n\nClone and start:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fnolanx-ai\u002Fnolanx.ai.git && cd nolanx.ai && .\u002Fdev.sh\n```\n\nOr, if you already cloned the repo:\n\n```bash\ncd nolanx.ai\n.\u002Fdev.sh\n```\n\nThis starts:\n\n- web: `http:\u002F\u002Flocalhost:3000`\n- api: `http:\u002F\u002Flocalhost:8080`\n- agent: `http:\u002F\u002F127.0.0.1:52178`\n\nThe browser opens `http:\u002F\u002Flocalhost:3000\u002Fnolanx` automatically after startup.\n\n## Development\n\n```bash\npnpm --dir apps\u002Fweb build\npnpm --dir apps\u002Fapi build\ncd apps\u002Fagent && .venv\u002Fbin\u002Fpython -m py_compile main.py\n```\n\n## Open-source runtime model\n\nThe stack is layered by capability.\n\n### Text\n\n- key: `OPENROUTER_API_KEY`\n- default model: `google\u002Fgemini-3.1-pro-preview`\n- responsibility: chat, planning, screenplay generation, routing, general reasoning\n- link: [OpenRouter](https:\u002F\u002Fopenrouter.ai\u002Fworkspaces\u002Fdefault\u002Fkeys)\n- self-host note: this is the first provider family you swap or extend\n\n### Image\n\n- key: `IMAGE_API_KEY`\n- default model: `openai\u002Fgpt-image-2`\n- responsibility: image generation, image editing, keyframes, visual references\n- link: [FAL](https:\u002F\u002Ffal.ai\u002Fdashboard\u002Fkeys)\n- self-host note: replace the image endpoint\u002Fprovider with your preferred image stack\n\n### Video\n\n- key: `VIDEO_API_KEY`\n- default model: `dreamina-seedance-2-0-260128`\n- default endpoint: `https:\u002F\u002Fnestapi.reelmind.ai\u002Fexternal-api\u002Fvideo\u002Fgenerate`\n- responsibility: shot generation\n- link: [ReelMind](https:\u002F\u002Freelmind.ai\u002Fplatform)\n- routing note: this is our own API platform and this routed path includes a 5% markup\n- self-host note: replace this layer with your preferred video provider or internal endpoint\n\n### Optional R2\n\n- keys: `R2_ACCOUNT_ID`, `R2_ACCESS_KEY_ID`, `R2_SECRET_ACCESS_KEY`, `R2_BUCKET_NAME`, `R2_PUBLIC_URL`\n- responsibility: uploaded references, generated assets, continuity-friendly persistent storage\n- link: [Cloudflare R2](https:\u002F\u002Fdash.cloudflare.com\u002F?to=\u002F:account\u002Fr2\u002Foverview)\n- note: NolanX still runs without it, but storage-backed flows become more limited\n\n### What happens if a key is missing\n\n- no `OPENROUTER_API_KEY`: chat, planning, and script generation do not run\n- no `IMAGE_API_KEY`: image generation and image editing do not run\n- no `VIDEO_API_KEY`: video generation does not run\n- no `R2_*`: NolanX still runs, but enhanced persistence \u002F upload \u002F continuity flows stay disabled\n\n## Configure runtime keys\n\n- Open `\u002Fnolanx` and click `Runtime Keys` for local overlays\n- `apps\u002Fagent\u002Fconfig.toml` is the primary self-host configuration file\n- `apps\u002Fapi\u002Fdata\u002Fruntime-config.json` stores UI-edited runtime values\n\n## Local data\n\nLocal data lives under `apps\u002Fapi\u002Fdata`:\n\n- `nolanx-db.json`\n- `runtime-config.json`\n\n\n## Credits\n\nNolanX was inspired, in a small way, by the Jaaz project about a year ago.\n\nOver that year we kept breaking through hard technical problems, modifying LangChain where needed, and learning from strong ideas in OpenClaw, Harness, Hermes, long-memory systems, and long-runtime agent design.\n\nThanks to the open-source community for the work that made this path possible.\n\nThis is our contribution back.\n\n## WhatsApp Group\n\nScan this QR code using the WhatsApp camera to join this group.\n\n![Nolanx.ai WhatsApp group QR code](apps\u002Fweb\u002Fpublic\u002Fwhatsapp.png)\n\n## Star History\n\n\u003Ca href=\"https:\u002F\u002Fwww.star-history.com\u002F?type=date&legend=top-left&repos=nolanx-ai%2Fnolanx.ai\">\n \u003Cpicture>\n   \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\"https:\u002F\u002Fapi.star-history.com\u002Fchart?repos=nolanx-ai\u002Fnolanx.ai&type=date&theme=dark&legend=top-left\" \u002F>\n   \u003Csource media=\"(prefers-color-scheme: light)\" srcset=\"https:\u002F\u002Fapi.star-history.com\u002Fchart?repos=nolanx-ai\u002Fnolanx.ai&type=date&legend=top-left\" \u002F>\n   \u003Cimg alt=\"Star History Chart\" src=\"https:\u002F\u002Fapi.star-history.com\u002Fchart?repos=nolanx-ai\u002Fnolanx.ai&type=date&legend=top-left\" \u002F>\n \u003C\u002Fpicture>\n\u003C\u002Fa>\n","NolanX 是一个开源的多模态代理基础设施，专为电影制作而设计。它通过集成多种模型（如Seedance 2.0, Kling 03等），将短视频生成器扩展成长达5分钟、30分钟甚至1小时的长视频生产管线，并在语言、图像、音频和视频之间保持创意的一致性。项目采用Python编写，具备强大的跨模态协调能力，能够处理从剧本逻辑到导演决策的整个创作流程。适用于需要长时间、高质量视频内容生成的场景，如电影、电视剧或复杂叙事广告的制作。MIT许可证下发布，既支持直接使用其在线服务，也允许用户自托管以满足特定需求。",2,"2026-06-11 03:53:35","CREATED_QUERY"]