[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-huggingface-space-demo-video":3,"mdc-mjhrzb-key":37,"related-repo-huggingface-space-demo-video":1655,"related-org-huggingface-space-demo-video":1665},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":27,"repoUrl":28,"updatedAt":29,"license":30,"forks":27,"topics":31,"repo":32,"sourceUrl":35,"mdContent":36},"space-demo-video","create demo videos for Hugging Face Spaces","Create a polished social demo video (MP4, 4:3, ~10s) for any Hugging Face Space with the space-demo-kit renderer. Use whenever the user asks for a demo, promo, showcase, or social video of a HF Space or model. Covers template selection (i2v, t2v, edit, avatar, a2a, gallery, vqa), manifest schema, live asset capture, rendering, verification, and delivery.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},"huggingface","Hugging Face","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fhuggingface.png",[12,16,18,21,24],{"name":13,"slug":14,"type":15},"Marketing","marketing","tag",{"name":9,"slug":17,"type":15},"hugging-face",{"name":19,"slug":20,"type":15},"Creative","creative",{"name":22,"slug":23,"type":15},"Hugging Face Spaces","hugging-face-spaces",{"name":25,"slug":26,"type":15},"Video","video",0,"https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fspace-demo-kit","2026-07-18T05:47:26.932972",null,[],{"repoUrl":28,"stars":27,"forks":27,"topics":33,"description":34},[],"Render polished Gradio-style demo videos for Hugging Face Spaces","https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fspace-demo-kit\u002Ftree\u002FHEAD\u002Fskills\u002Fspace-demo-video","---\nname: space-demo-video\ndescription: \"Create a polished social demo video (MP4, 4:3, ~10s) for any Hugging Face Space with the space-demo-kit renderer. Use whenever the user asks for a demo, promo, showcase, or social video of a HF Space or model. Covers template selection (i2v, t2v, edit, avatar, a2a, gallery, vqa), manifest schema, live asset capture, rendering, verification, and delivery.\"\n---\n\n# Space Demo Video (space-demo-kit)\n\nThe kit renders a Gradio-style \"app window\" video: a hero intro where the input opens\nnear full-frame with the prompt typing in large text (readable at mobile feed size),\na morph into the app layout with the Space's real header (emoji, author\u002Fname, likes,\nhardware), a Generate press, a progress sweep, and the real output filling the output\npanel edge-to-edge. Text auto-fits its panels (short answers render huge, long ones stay\nreadable). Branding is extracted from the Space page automatically. Output: 1440x1080\n(4:3), 30 fps, ~12–17 s, H.264, silent unless the task itself involves audio.\n\nKit location on this machine: `\u002Froot\u002Fcode\u002Fspace-demo-kit`\nRender command (the ONLY way to produce the video):\n\n```bash\ncd \u002Froot\u002Fcode\u002Fspace-demo-kit && python3 render.py manifests\u002F\u003Cname>.json\n# -> out\u002F\u003Cname>\u002F\u003Cname>.mp4  +  out\u002F\u003Cname>\u002F\u003Cname>-contact.jpg\n```\n\n## Golden rules\n\n1. **Never design video frames by hand.** No PIL cards, no ffmpeg drawtext\u002Foverlay design,\n   no custom HTML. The kit owns 100% of the design; you only supply facts and assets via a\n   manifest JSON.\n2. **Real outputs.** Generate the output by calling the live Space (`gradio_client`) with a\n   real input. Only mock when live inference is impossible (quota, broken Space), and then\n   label it honestly in the manifest `meta` field (e.g. `\"mock demo assets\"`) and tell the user.\n3. **Real branding.** Pull `emoji`, `colorFrom`, `colorTo`, `likes`, and hardware from\n   `https:\u002F\u002Fhuggingface.co\u002Fapi\u002Fspaces\u002F\u003Cid>` (`cardData` + `runtime.hardware.current`).\n   If the user gives a MODEL id, resolve its demo Space first (model page \u002F `spaces` field of\n   `https:\u002F\u002Fhuggingface.co\u002Fapi\u002Fmodels\u002F\u003Cid>`).\n4. **Prompts under 160 chars.** The renderer truncates displayed prompts around 150 chars;\n   pick or write inputs that fit.\n5. **Verify before delivering**: `ffprobe` the MP4 (duration, 1440x1080) and look at the\n   contact sheet (no overflowing text, output visible). Then send the MP4 to the user as media.\n6. **One look only.** Never put a `\"style\"` key in a manifest — the app-window template is the\n   only production design and the renderer ignores styles. Anything under `experiments\u002F` is a\n   rejected prototype: never use it as an example or reference.\n\n## Asset quality bar (hard requirements)\n\n- **Inputs must be real content**: a real photograph, illustration, or clip at least 512 px on\n  the short side. NEVER an emoji, icon, favicon, logo, or clipart — tiny images upscale into\n  blurry blobs and ruin the video. Verify every asset with `file` + dimensions before using it.\n- **Best input source**: the Space's own example assets (the `examples` in its Gradio config,\n  or files in the Space repo via `https:\u002F\u002Fhuggingface.co\u002Fapi\u002Fspaces\u002F\u003Cid>\u002Ftree\u002Fmain`). They are\n  known to work with the model.\n- **Before\u002Fafter coherence (`edit`, `a2a`)**: the output MUST actually be derived from the\n  input. Never pair two unrelated assets — an incoherent before\u002Fafter is worse than no video.\n  If live inference fails: retry via `gradio_client` (not hand-built HTTP calls), try another\n  example input, and if it still fails, report the failure to the user and STOP. Do not\n  fabricate the pair.\n\n## Template selection\n\n| template | Space task | required manifest fields |\n|---|---|---|\n| `i2v` | image → video | `input_image`, `prompt`, `video` (generated clip) |\n| `t2v` | text → video | `prompt`, `video` |\n| `edit` | image + instruction → image | `input_image`, `prompt`, `before_image`, `after_image` |\n| `avatar` | portrait + audio → talking video | `input_image`, `audio`, `video` (audio can be the generated mp4 itself) |\n| `a2a` | audio → audio | `audio_in`, `audio_out` |\n| `gallery` | one input → several images | `input_image`, `prompt`, `outputs` (list of `{path, label}`, 4 items) |\n| `vqa` | video\u002Fimage + question → text | `video`, `prompt`, `answer` |\n\nText-to-image Spaces use `t2v`: turn the generated image into a short static clip first\n(`ffmpeg -loop 1 -i out.webp -t 6 -r 30 -pix_fmt yuv420p static.mp4`) and pass it as `video`\n(see `manifests\u002Fz-image-turbo-demo.json`). There is no `t2i` template.\n\n## Manifest schema\n\nCommon fields (all templates):\n\n```json\n{\n  \"name\": \"my-demo\",                    \u002F\u002F output folder + file name\n  \"template\": \"edit\",\n  \"space\": {\n    \"id\": \"owner\u002Fspace-name\",\n    \"emoji\": \"🧬\",                      \u002F\u002F from cardData\n    \"colorFrom\": \"blue\",                \u002F\u002F from cardData\n    \"colorTo\": \"gray\",                  \u002F\u002F from cardData\n    \"likes\": 27,                        \u002F\u002F from the API\n    \"hardware\": \"zero-a10g\"             \u002F\u002F runtime.hardware.current\n  },\n  \"meta\": \"live output · zero-a10g\",    \u002F\u002F small honest footer note; use \"mock demo assets\" for mocks\n  \"button\": \"Generate\",                 \u002F\u002F optional; e.g. \"Enhance\", \"Animate\"\n  \"input_label\": \"source image\",        \u002F\u002F optional label overrides\n  \"prompt_label\": \"edit instruction\",\n  \"output_label\": \"edited image\",\n  \"audio_gain\": 2.0,                    \u002F\u002F optional; default 2.0 (with limiter)\n  \"hero\": false                         \u002F\u002F optional; set false to skip the full-frame intro\n}\n```\n\nPlus the template-specific fields from the table. All asset paths are absolute paths on\nthis machine. Store working assets under `\u002Froot\u002Fcode\u002F\u003Cproject-dir>\u002F` as usual.\n\n## Workflow\n\n1. `curl -s https:\u002F\u002Fhuggingface.co\u002Fapi\u002Fspaces\u002F\u003Cid>` → metadata for `space{}` fields.\n2. Pick the template from the table. If the Space's task fits none of them, say so and pick\n   the closest; do NOT invent a new design.\n3. Capture assets: choose a good input (the Space's own example assets are ideal), run the\n   live Space with `gradio_client`, download the real output. Verify the output file with\n   `ffprobe`\u002F`file` before rendering.\n4. Write `manifests\u002F\u003Cname>.json` in the kit.\n5. Render (command above). First run downloads fonts\u002Femoji (needs network) and, for video\n   inputs, extracts frames. A render takes roughly 3–5 minutes — do not assume it hung.\n6. Verify MP4 + contact sheet, then deliver the MP4 to the user as a media message.\n\n## Codex dispatch (OpenClaw — mandatory)\n\nDemo-video requests are heavy (live inference + a ~2 min render). Do NOT run inference or\nrendering inline in the main session — it blocks you for many minutes. ALWAYS dispatch a\ncodex session per the codex-monitor skill (setsid, `\u003C\u002Fdev\u002Fnull`, CODEX_HOME pinned) and let\nthe CODEX_FINISHED wake drive verification + delivery. Include in the codex prompt:\n\"Read \u002Froot\u002Fcode\u002Fspace-demo-kit\u002Fskills\u002Fspace-demo-video\u002FSKILL.md and follow it exactly.\nUse the kit's render.py for ALL visuals; do not design any frame yourself.\"\n\n## Gotchas\n\n- Long Space ids auto-shrink in the footer; long repo names get ellipsized in the header — fine.\n- The renderer needs Playwright + Chromium (already installed system-wide here).\n- Audio templates (`avatar`, `a2a`) mux the real audio automatically, delayed to the moment\n  playback starts on screen, at 2x gain with a limiter.\n- `gallery` wants exactly 4 outputs for the 2x2 grid.\n- `vqa` answers auto-fit the output panel: keep them 30–60 words so the type stays large.\n- Do not edit `templates\u002Fapp-window.html` or `render.py` for a one-off video. If a Space\n  genuinely needs a new mode, report that instead of hacking around the kit.\n",{"data":38,"body":39},{"name":4,"description":6},{"type":40,"children":41},"root",[42,51,57,71,158,165,349,355,435,441,748,791,797,802,1457,1470,1476,1552,1558,1571,1577,1649],{"type":43,"tag":44,"props":45,"children":47},"element","h1",{"id":46},"space-demo-video-space-demo-kit",[48],{"type":49,"value":50},"text","Space Demo Video (space-demo-kit)",{"type":43,"tag":52,"props":53,"children":54},"p",{},[55],{"type":49,"value":56},"The kit renders a Gradio-style \"app window\" video: a hero intro where the input opens\nnear full-frame with the prompt typing in large text (readable at mobile feed size),\na morph into the app layout with the Space's real header (emoji, author\u002Fname, likes,\nhardware), a Generate press, a progress sweep, and the real output filling the output\npanel edge-to-edge. Text auto-fits its panels (short answers render huge, long ones stay\nreadable). Branding is extracted from the Space page automatically. Output: 1440x1080\n(4:3), 30 fps, ~12–17 s, H.264, silent unless the task itself involves audio.",{"type":43,"tag":52,"props":58,"children":59},{},[60,62,69],{"type":49,"value":61},"Kit location on this machine: ",{"type":43,"tag":63,"props":64,"children":66},"code",{"className":65},[],[67],{"type":49,"value":68},"\u002Froot\u002Fcode\u002Fspace-demo-kit",{"type":49,"value":70},"\nRender command (the ONLY way to produce the video):",{"type":43,"tag":72,"props":73,"children":78},"pre",{"className":74,"code":75,"language":76,"meta":77,"style":77},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","cd \u002Froot\u002Fcode\u002Fspace-demo-kit && python3 render.py manifests\u002F\u003Cname>.json\n# -> out\u002F\u003Cname>\u002F\u003Cname>.mp4  +  out\u002F\u003Cname>\u002F\u003Cname>-contact.jpg\n","bash","",[79],{"type":43,"tag":63,"props":80,"children":81},{"__ignoreMap":77},[82,148],{"type":43,"tag":83,"props":84,"children":87},"span",{"class":85,"line":86},"line",1,[88,94,100,106,112,117,122,127,132,138,143],{"type":43,"tag":83,"props":89,"children":91},{"style":90},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[92],{"type":49,"value":93},"cd",{"type":43,"tag":83,"props":95,"children":97},{"style":96},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[98],{"type":49,"value":99}," \u002Froot\u002Fcode\u002Fspace-demo-kit",{"type":43,"tag":83,"props":101,"children":103},{"style":102},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[104],{"type":49,"value":105}," &&",{"type":43,"tag":83,"props":107,"children":109},{"style":108},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[110],{"type":49,"value":111}," python3",{"type":43,"tag":83,"props":113,"children":114},{"style":96},[115],{"type":49,"value":116}," render.py",{"type":43,"tag":83,"props":118,"children":119},{"style":96},[120],{"type":49,"value":121}," manifests\u002F",{"type":43,"tag":83,"props":123,"children":124},{"style":102},[125],{"type":49,"value":126},"\u003C",{"type":43,"tag":83,"props":128,"children":129},{"style":96},[130],{"type":49,"value":131},"nam",{"type":43,"tag":83,"props":133,"children":135},{"style":134},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[136],{"type":49,"value":137},"e",{"type":43,"tag":83,"props":139,"children":140},{"style":102},[141],{"type":49,"value":142},">",{"type":43,"tag":83,"props":144,"children":145},{"style":96},[146],{"type":49,"value":147},".json\n",{"type":43,"tag":83,"props":149,"children":151},{"class":85,"line":150},2,[152],{"type":43,"tag":83,"props":153,"children":155},{"style":154},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[156],{"type":49,"value":157},"# -> out\u002F\u003Cname>\u002F\u003Cname>.mp4  +  out\u002F\u003Cname>\u002F\u003Cname>-contact.jpg\n",{"type":43,"tag":159,"props":160,"children":162},"h2",{"id":161},"golden-rules",[163],{"type":49,"value":164},"Golden rules",{"type":43,"tag":166,"props":167,"children":168},"ol",{},[169,181,215,295,305,323],{"type":43,"tag":170,"props":171,"children":172},"li",{},[173,179],{"type":43,"tag":174,"props":175,"children":176},"strong",{},[177],{"type":49,"value":178},"Never design video frames by hand.",{"type":49,"value":180}," No PIL cards, no ffmpeg drawtext\u002Foverlay design,\nno custom HTML. The kit owns 100% of the design; you only supply facts and assets via a\nmanifest JSON.",{"type":43,"tag":170,"props":182,"children":183},{},[184,189,191,197,199,205,207,213],{"type":43,"tag":174,"props":185,"children":186},{},[187],{"type":49,"value":188},"Real outputs.",{"type":49,"value":190}," Generate the output by calling the live Space (",{"type":43,"tag":63,"props":192,"children":194},{"className":193},[],[195],{"type":49,"value":196},"gradio_client",{"type":49,"value":198},") with a\nreal input. Only mock when live inference is impossible (quota, broken Space), and then\nlabel it honestly in the manifest ",{"type":43,"tag":63,"props":200,"children":202},{"className":201},[],[203],{"type":49,"value":204},"meta",{"type":49,"value":206}," field (e.g. ",{"type":43,"tag":63,"props":208,"children":210},{"className":209},[],[211],{"type":49,"value":212},"\"mock demo assets\"",{"type":49,"value":214},") and tell the user.",{"type":43,"tag":170,"props":216,"children":217},{},[218,223,225,231,233,239,240,246,247,253,255,261,263,269,271,277,279,285,287,293],{"type":43,"tag":174,"props":219,"children":220},{},[221],{"type":49,"value":222},"Real branding.",{"type":49,"value":224}," Pull ",{"type":43,"tag":63,"props":226,"children":228},{"className":227},[],[229],{"type":49,"value":230},"emoji",{"type":49,"value":232},", ",{"type":43,"tag":63,"props":234,"children":236},{"className":235},[],[237],{"type":49,"value":238},"colorFrom",{"type":49,"value":232},{"type":43,"tag":63,"props":241,"children":243},{"className":242},[],[244],{"type":49,"value":245},"colorTo",{"type":49,"value":232},{"type":43,"tag":63,"props":248,"children":250},{"className":249},[],[251],{"type":49,"value":252},"likes",{"type":49,"value":254},", and hardware from\n",{"type":43,"tag":63,"props":256,"children":258},{"className":257},[],[259],{"type":49,"value":260},"https:\u002F\u002Fhuggingface.co\u002Fapi\u002Fspaces\u002F\u003Cid>",{"type":49,"value":262}," (",{"type":43,"tag":63,"props":264,"children":266},{"className":265},[],[267],{"type":49,"value":268},"cardData",{"type":49,"value":270}," + ",{"type":43,"tag":63,"props":272,"children":274},{"className":273},[],[275],{"type":49,"value":276},"runtime.hardware.current",{"type":49,"value":278},").\nIf the user gives a MODEL id, resolve its demo Space first (model page \u002F ",{"type":43,"tag":63,"props":280,"children":282},{"className":281},[],[283],{"type":49,"value":284},"spaces",{"type":49,"value":286}," field of\n",{"type":43,"tag":63,"props":288,"children":290},{"className":289},[],[291],{"type":49,"value":292},"https:\u002F\u002Fhuggingface.co\u002Fapi\u002Fmodels\u002F\u003Cid>",{"type":49,"value":294},").",{"type":43,"tag":170,"props":296,"children":297},{},[298,303],{"type":43,"tag":174,"props":299,"children":300},{},[301],{"type":49,"value":302},"Prompts under 160 chars.",{"type":49,"value":304}," The renderer truncates displayed prompts around 150 chars;\npick or write inputs that fit.",{"type":43,"tag":170,"props":306,"children":307},{},[308,313,315,321],{"type":43,"tag":174,"props":309,"children":310},{},[311],{"type":49,"value":312},"Verify before delivering",{"type":49,"value":314},": ",{"type":43,"tag":63,"props":316,"children":318},{"className":317},[],[319],{"type":49,"value":320},"ffprobe",{"type":49,"value":322}," the MP4 (duration, 1440x1080) and look at the\ncontact sheet (no overflowing text, output visible). Then send the MP4 to the user as media.",{"type":43,"tag":170,"props":324,"children":325},{},[326,331,333,339,341,347],{"type":43,"tag":174,"props":327,"children":328},{},[329],{"type":49,"value":330},"One look only.",{"type":49,"value":332}," Never put a ",{"type":43,"tag":63,"props":334,"children":336},{"className":335},[],[337],{"type":49,"value":338},"\"style\"",{"type":49,"value":340}," key in a manifest — the app-window template is the\nonly production design and the renderer ignores styles. Anything under ",{"type":43,"tag":63,"props":342,"children":344},{"className":343},[],[345],{"type":49,"value":346},"experiments\u002F",{"type":49,"value":348}," is a\nrejected prototype: never use it as an example or reference.",{"type":43,"tag":159,"props":350,"children":352},{"id":351},"asset-quality-bar-hard-requirements",[353],{"type":49,"value":354},"Asset quality bar (hard requirements)",{"type":43,"tag":356,"props":357,"children":358},"ul",{},[359,377,403],{"type":43,"tag":170,"props":360,"children":361},{},[362,367,369,375],{"type":43,"tag":174,"props":363,"children":364},{},[365],{"type":49,"value":366},"Inputs must be real content",{"type":49,"value":368},": a real photograph, illustration, or clip at least 512 px on\nthe short side. NEVER an emoji, icon, favicon, logo, or clipart — tiny images upscale into\nblurry blobs and ruin the video. Verify every asset with ",{"type":43,"tag":63,"props":370,"children":372},{"className":371},[],[373],{"type":49,"value":374},"file",{"type":49,"value":376}," + dimensions before using it.",{"type":43,"tag":170,"props":378,"children":379},{},[380,385,387,393,395,401],{"type":43,"tag":174,"props":381,"children":382},{},[383],{"type":49,"value":384},"Best input source",{"type":49,"value":386},": the Space's own example assets (the ",{"type":43,"tag":63,"props":388,"children":390},{"className":389},[],[391],{"type":49,"value":392},"examples",{"type":49,"value":394}," in its Gradio config,\nor files in the Space repo via ",{"type":43,"tag":63,"props":396,"children":398},{"className":397},[],[399],{"type":49,"value":400},"https:\u002F\u002Fhuggingface.co\u002Fapi\u002Fspaces\u002F\u003Cid>\u002Ftree\u002Fmain",{"type":49,"value":402},"). They are\nknown to work with the model.",{"type":43,"tag":170,"props":404,"children":405},{},[406,426,428,433],{"type":43,"tag":174,"props":407,"children":408},{},[409,411,417,418,424],{"type":49,"value":410},"Before\u002Fafter coherence (",{"type":43,"tag":63,"props":412,"children":414},{"className":413},[],[415],{"type":49,"value":416},"edit",{"type":49,"value":232},{"type":43,"tag":63,"props":419,"children":421},{"className":420},[],[422],{"type":49,"value":423},"a2a",{"type":49,"value":425},")",{"type":49,"value":427},": the output MUST actually be derived from the\ninput. Never pair two unrelated assets — an incoherent before\u002Fafter is worse than no video.\nIf live inference fails: retry via ",{"type":43,"tag":63,"props":429,"children":431},{"className":430},[],[432],{"type":49,"value":196},{"type":49,"value":434}," (not hand-built HTTP calls), try another\nexample input, and if it still fails, report the failure to the user and STOP. Do not\nfabricate the pair.",{"type":43,"tag":159,"props":436,"children":438},{"id":437},"template-selection",[439],{"type":49,"value":440},"Template selection",{"type":43,"tag":442,"props":443,"children":444},"table",{},[445,469],{"type":43,"tag":446,"props":447,"children":448},"thead",{},[449],{"type":43,"tag":450,"props":451,"children":452},"tr",{},[453,459,464],{"type":43,"tag":454,"props":455,"children":456},"th",{},[457],{"type":49,"value":458},"template",{"type":43,"tag":454,"props":460,"children":461},{},[462],{"type":49,"value":463},"Space task",{"type":43,"tag":454,"props":465,"children":466},{},[467],{"type":49,"value":468},"required manifest fields",{"type":43,"tag":470,"props":471,"children":472},"tbody",{},[473,515,546,590,630,662,710],{"type":43,"tag":450,"props":474,"children":475},{},[476,486,491],{"type":43,"tag":477,"props":478,"children":479},"td",{},[480],{"type":43,"tag":63,"props":481,"children":483},{"className":482},[],[484],{"type":49,"value":485},"i2v",{"type":43,"tag":477,"props":487,"children":488},{},[489],{"type":49,"value":490},"image → video",{"type":43,"tag":477,"props":492,"children":493},{},[494,500,501,507,508,513],{"type":43,"tag":63,"props":495,"children":497},{"className":496},[],[498],{"type":49,"value":499},"input_image",{"type":49,"value":232},{"type":43,"tag":63,"props":502,"children":504},{"className":503},[],[505],{"type":49,"value":506},"prompt",{"type":49,"value":232},{"type":43,"tag":63,"props":509,"children":511},{"className":510},[],[512],{"type":49,"value":26},{"type":49,"value":514}," (generated clip)",{"type":43,"tag":450,"props":516,"children":517},{},[518,527,532],{"type":43,"tag":477,"props":519,"children":520},{},[521],{"type":43,"tag":63,"props":522,"children":524},{"className":523},[],[525],{"type":49,"value":526},"t2v",{"type":43,"tag":477,"props":528,"children":529},{},[530],{"type":49,"value":531},"text → video",{"type":43,"tag":477,"props":533,"children":534},{},[535,540,541],{"type":43,"tag":63,"props":536,"children":538},{"className":537},[],[539],{"type":49,"value":506},{"type":49,"value":232},{"type":43,"tag":63,"props":542,"children":544},{"className":543},[],[545],{"type":49,"value":26},{"type":43,"tag":450,"props":547,"children":548},{},[549,557,562],{"type":43,"tag":477,"props":550,"children":551},{},[552],{"type":43,"tag":63,"props":553,"children":555},{"className":554},[],[556],{"type":49,"value":416},{"type":43,"tag":477,"props":558,"children":559},{},[560],{"type":49,"value":561},"image + instruction → image",{"type":43,"tag":477,"props":563,"children":564},{},[565,570,571,576,577,583,584],{"type":43,"tag":63,"props":566,"children":568},{"className":567},[],[569],{"type":49,"value":499},{"type":49,"value":232},{"type":43,"tag":63,"props":572,"children":574},{"className":573},[],[575],{"type":49,"value":506},{"type":49,"value":232},{"type":43,"tag":63,"props":578,"children":580},{"className":579},[],[581],{"type":49,"value":582},"before_image",{"type":49,"value":232},{"type":43,"tag":63,"props":585,"children":587},{"className":586},[],[588],{"type":49,"value":589},"after_image",{"type":43,"tag":450,"props":591,"children":592},{},[593,602,607],{"type":43,"tag":477,"props":594,"children":595},{},[596],{"type":43,"tag":63,"props":597,"children":599},{"className":598},[],[600],{"type":49,"value":601},"avatar",{"type":43,"tag":477,"props":603,"children":604},{},[605],{"type":49,"value":606},"portrait + audio → talking video",{"type":43,"tag":477,"props":608,"children":609},{},[610,615,616,622,623,628],{"type":43,"tag":63,"props":611,"children":613},{"className":612},[],[614],{"type":49,"value":499},{"type":49,"value":232},{"type":43,"tag":63,"props":617,"children":619},{"className":618},[],[620],{"type":49,"value":621},"audio",{"type":49,"value":232},{"type":43,"tag":63,"props":624,"children":626},{"className":625},[],[627],{"type":49,"value":26},{"type":49,"value":629}," (audio can be the generated mp4 itself)",{"type":43,"tag":450,"props":631,"children":632},{},[633,641,646],{"type":43,"tag":477,"props":634,"children":635},{},[636],{"type":43,"tag":63,"props":637,"children":639},{"className":638},[],[640],{"type":49,"value":423},{"type":43,"tag":477,"props":642,"children":643},{},[644],{"type":49,"value":645},"audio → audio",{"type":43,"tag":477,"props":647,"children":648},{},[649,655,656],{"type":43,"tag":63,"props":650,"children":652},{"className":651},[],[653],{"type":49,"value":654},"audio_in",{"type":49,"value":232},{"type":43,"tag":63,"props":657,"children":659},{"className":658},[],[660],{"type":49,"value":661},"audio_out",{"type":43,"tag":450,"props":663,"children":664},{},[665,674,679],{"type":43,"tag":477,"props":666,"children":667},{},[668],{"type":43,"tag":63,"props":669,"children":671},{"className":670},[],[672],{"type":49,"value":673},"gallery",{"type":43,"tag":477,"props":675,"children":676},{},[677],{"type":49,"value":678},"one input → several images",{"type":43,"tag":477,"props":680,"children":681},{},[682,687,688,693,694,700,702,708],{"type":43,"tag":63,"props":683,"children":685},{"className":684},[],[686],{"type":49,"value":499},{"type":49,"value":232},{"type":43,"tag":63,"props":689,"children":691},{"className":690},[],[692],{"type":49,"value":506},{"type":49,"value":232},{"type":43,"tag":63,"props":695,"children":697},{"className":696},[],[698],{"type":49,"value":699},"outputs",{"type":49,"value":701}," (list of ",{"type":43,"tag":63,"props":703,"children":705},{"className":704},[],[706],{"type":49,"value":707},"{path, label}",{"type":49,"value":709},", 4 items)",{"type":43,"tag":450,"props":711,"children":712},{},[713,722,727],{"type":43,"tag":477,"props":714,"children":715},{},[716],{"type":43,"tag":63,"props":717,"children":719},{"className":718},[],[720],{"type":49,"value":721},"vqa",{"type":43,"tag":477,"props":723,"children":724},{},[725],{"type":49,"value":726},"video\u002Fimage + question → text",{"type":43,"tag":477,"props":728,"children":729},{},[730,735,736,741,742],{"type":43,"tag":63,"props":731,"children":733},{"className":732},[],[734],{"type":49,"value":26},{"type":49,"value":232},{"type":43,"tag":63,"props":737,"children":739},{"className":738},[],[740],{"type":49,"value":506},{"type":49,"value":232},{"type":43,"tag":63,"props":743,"children":745},{"className":744},[],[746],{"type":49,"value":747},"answer",{"type":43,"tag":52,"props":749,"children":750},{},[751,753,758,760,766,768,773,775,781,783,789],{"type":49,"value":752},"Text-to-image Spaces use ",{"type":43,"tag":63,"props":754,"children":756},{"className":755},[],[757],{"type":49,"value":526},{"type":49,"value":759},": turn the generated image into a short static clip first\n(",{"type":43,"tag":63,"props":761,"children":763},{"className":762},[],[764],{"type":49,"value":765},"ffmpeg -loop 1 -i out.webp -t 6 -r 30 -pix_fmt yuv420p static.mp4",{"type":49,"value":767},") and pass it as ",{"type":43,"tag":63,"props":769,"children":771},{"className":770},[],[772],{"type":49,"value":26},{"type":49,"value":774},"\n(see ",{"type":43,"tag":63,"props":776,"children":778},{"className":777},[],[779],{"type":49,"value":780},"manifests\u002Fz-image-turbo-demo.json",{"type":49,"value":782},"). There is no ",{"type":43,"tag":63,"props":784,"children":786},{"className":785},[],[787],{"type":49,"value":788},"t2i",{"type":49,"value":790}," template.",{"type":43,"tag":159,"props":792,"children":794},{"id":793},"manifest-schema",[795],{"type":49,"value":796},"Manifest schema",{"type":43,"tag":52,"props":798,"children":799},{},[800],{"type":49,"value":801},"Common fields (all templates):",{"type":43,"tag":72,"props":803,"children":807},{"className":804,"code":805,"language":806,"meta":77,"style":77},"language-json shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","{\n  \"name\": \"my-demo\",                    \u002F\u002F output folder + file name\n  \"template\": \"edit\",\n  \"space\": {\n    \"id\": \"owner\u002Fspace-name\",\n    \"emoji\": \"🧬\",                      \u002F\u002F from cardData\n    \"colorFrom\": \"blue\",                \u002F\u002F from cardData\n    \"colorTo\": \"gray\",                  \u002F\u002F from cardData\n    \"likes\": 27,                        \u002F\u002F from the API\n    \"hardware\": \"zero-a10g\"             \u002F\u002F runtime.hardware.current\n  },\n  \"meta\": \"live output · zero-a10g\",    \u002F\u002F small honest footer note; use \"mock demo assets\" for mocks\n  \"button\": \"Generate\",                 \u002F\u002F optional; e.g. \"Enhance\", \"Animate\"\n  \"input_label\": \"source image\",        \u002F\u002F optional label overrides\n  \"prompt_label\": \"edit instruction\",\n  \"output_label\": \"edited image\",\n  \"audio_gain\": 2.0,                    \u002F\u002F optional; default 2.0 (with limiter)\n  \"hero\": false                         \u002F\u002F optional; set false to skip the full-frame intro\n}\n","json",[808],{"type":43,"tag":63,"props":809,"children":810},{"__ignoreMap":77},[811,819,867,904,930,969,1011,1053,1095,1130,1169,1178,1220,1263,1306,1344,1382,1417,1448],{"type":43,"tag":83,"props":812,"children":813},{"class":85,"line":86},[814],{"type":43,"tag":83,"props":815,"children":816},{"style":102},[817],{"type":49,"value":818},"{\n",{"type":43,"tag":83,"props":820,"children":821},{"class":85,"line":150},[822,827,833,838,843,848,853,857,862],{"type":43,"tag":83,"props":823,"children":824},{"style":102},[825],{"type":49,"value":826},"  \"",{"type":43,"tag":83,"props":828,"children":830},{"style":829},"--shiki-light:#9C3EDA;--shiki-default:#C792EA;--shiki-dark:#C792EA",[831],{"type":49,"value":832},"name",{"type":43,"tag":83,"props":834,"children":835},{"style":102},[836],{"type":49,"value":837},"\"",{"type":43,"tag":83,"props":839,"children":840},{"style":102},[841],{"type":49,"value":842},":",{"type":43,"tag":83,"props":844,"children":845},{"style":102},[846],{"type":49,"value":847}," \"",{"type":43,"tag":83,"props":849,"children":850},{"style":96},[851],{"type":49,"value":852},"my-demo",{"type":43,"tag":83,"props":854,"children":855},{"style":102},[856],{"type":49,"value":837},{"type":43,"tag":83,"props":858,"children":859},{"style":102},[860],{"type":49,"value":861},",",{"type":43,"tag":83,"props":863,"children":864},{"style":154},[865],{"type":49,"value":866},"                    \u002F\u002F output folder + file name\n",{"type":43,"tag":83,"props":868,"children":870},{"class":85,"line":869},3,[871,875,879,883,887,891,895,899],{"type":43,"tag":83,"props":872,"children":873},{"style":102},[874],{"type":49,"value":826},{"type":43,"tag":83,"props":876,"children":877},{"style":829},[878],{"type":49,"value":458},{"type":43,"tag":83,"props":880,"children":881},{"style":102},[882],{"type":49,"value":837},{"type":43,"tag":83,"props":884,"children":885},{"style":102},[886],{"type":49,"value":842},{"type":43,"tag":83,"props":888,"children":889},{"style":102},[890],{"type":49,"value":847},{"type":43,"tag":83,"props":892,"children":893},{"style":96},[894],{"type":49,"value":416},{"type":43,"tag":83,"props":896,"children":897},{"style":102},[898],{"type":49,"value":837},{"type":43,"tag":83,"props":900,"children":901},{"style":102},[902],{"type":49,"value":903},",\n",{"type":43,"tag":83,"props":905,"children":907},{"class":85,"line":906},4,[908,912,917,921,925],{"type":43,"tag":83,"props":909,"children":910},{"style":102},[911],{"type":49,"value":826},{"type":43,"tag":83,"props":913,"children":914},{"style":829},[915],{"type":49,"value":916},"space",{"type":43,"tag":83,"props":918,"children":919},{"style":102},[920],{"type":49,"value":837},{"type":43,"tag":83,"props":922,"children":923},{"style":102},[924],{"type":49,"value":842},{"type":43,"tag":83,"props":926,"children":927},{"style":102},[928],{"type":49,"value":929}," {\n",{"type":43,"tag":83,"props":931,"children":933},{"class":85,"line":932},5,[934,939,944,948,952,956,961,965],{"type":43,"tag":83,"props":935,"children":936},{"style":102},[937],{"type":49,"value":938},"    \"",{"type":43,"tag":83,"props":940,"children":941},{"style":108},[942],{"type":49,"value":943},"id",{"type":43,"tag":83,"props":945,"children":946},{"style":102},[947],{"type":49,"value":837},{"type":43,"tag":83,"props":949,"children":950},{"style":102},[951],{"type":49,"value":842},{"type":43,"tag":83,"props":953,"children":954},{"style":102},[955],{"type":49,"value":847},{"type":43,"tag":83,"props":957,"children":958},{"style":96},[959],{"type":49,"value":960},"owner\u002Fspace-name",{"type":43,"tag":83,"props":962,"children":963},{"style":102},[964],{"type":49,"value":837},{"type":43,"tag":83,"props":966,"children":967},{"style":102},[968],{"type":49,"value":903},{"type":43,"tag":83,"props":970,"children":972},{"class":85,"line":971},6,[973,977,981,985,989,993,998,1002,1006],{"type":43,"tag":83,"props":974,"children":975},{"style":102},[976],{"type":49,"value":938},{"type":43,"tag":83,"props":978,"children":979},{"style":108},[980],{"type":49,"value":230},{"type":43,"tag":83,"props":982,"children":983},{"style":102},[984],{"type":49,"value":837},{"type":43,"tag":83,"props":986,"children":987},{"style":102},[988],{"type":49,"value":842},{"type":43,"tag":83,"props":990,"children":991},{"style":102},[992],{"type":49,"value":847},{"type":43,"tag":83,"props":994,"children":995},{"style":96},[996],{"type":49,"value":997},"🧬",{"type":43,"tag":83,"props":999,"children":1000},{"style":102},[1001],{"type":49,"value":837},{"type":43,"tag":83,"props":1003,"children":1004},{"style":102},[1005],{"type":49,"value":861},{"type":43,"tag":83,"props":1007,"children":1008},{"style":154},[1009],{"type":49,"value":1010},"                      \u002F\u002F from cardData\n",{"type":43,"tag":83,"props":1012,"children":1014},{"class":85,"line":1013},7,[1015,1019,1023,1027,1031,1035,1040,1044,1048],{"type":43,"tag":83,"props":1016,"children":1017},{"style":102},[1018],{"type":49,"value":938},{"type":43,"tag":83,"props":1020,"children":1021},{"style":108},[1022],{"type":49,"value":238},{"type":43,"tag":83,"props":1024,"children":1025},{"style":102},[1026],{"type":49,"value":837},{"type":43,"tag":83,"props":1028,"children":1029},{"style":102},[1030],{"type":49,"value":842},{"type":43,"tag":83,"props":1032,"children":1033},{"style":102},[1034],{"type":49,"value":847},{"type":43,"tag":83,"props":1036,"children":1037},{"style":96},[1038],{"type":49,"value":1039},"blue",{"type":43,"tag":83,"props":1041,"children":1042},{"style":102},[1043],{"type":49,"value":837},{"type":43,"tag":83,"props":1045,"children":1046},{"style":102},[1047],{"type":49,"value":861},{"type":43,"tag":83,"props":1049,"children":1050},{"style":154},[1051],{"type":49,"value":1052},"                \u002F\u002F from cardData\n",{"type":43,"tag":83,"props":1054,"children":1056},{"class":85,"line":1055},8,[1057,1061,1065,1069,1073,1077,1082,1086,1090],{"type":43,"tag":83,"props":1058,"children":1059},{"style":102},[1060],{"type":49,"value":938},{"type":43,"tag":83,"props":1062,"children":1063},{"style":108},[1064],{"type":49,"value":245},{"type":43,"tag":83,"props":1066,"children":1067},{"style":102},[1068],{"type":49,"value":837},{"type":43,"tag":83,"props":1070,"children":1071},{"style":102},[1072],{"type":49,"value":842},{"type":43,"tag":83,"props":1074,"children":1075},{"style":102},[1076],{"type":49,"value":847},{"type":43,"tag":83,"props":1078,"children":1079},{"style":96},[1080],{"type":49,"value":1081},"gray",{"type":43,"tag":83,"props":1083,"children":1084},{"style":102},[1085],{"type":49,"value":837},{"type":43,"tag":83,"props":1087,"children":1088},{"style":102},[1089],{"type":49,"value":861},{"type":43,"tag":83,"props":1091,"children":1092},{"style":154},[1093],{"type":49,"value":1094},"                  \u002F\u002F from cardData\n",{"type":43,"tag":83,"props":1096,"children":1098},{"class":85,"line":1097},9,[1099,1103,1107,1111,1115,1121,1125],{"type":43,"tag":83,"props":1100,"children":1101},{"style":102},[1102],{"type":49,"value":938},{"type":43,"tag":83,"props":1104,"children":1105},{"style":108},[1106],{"type":49,"value":252},{"type":43,"tag":83,"props":1108,"children":1109},{"style":102},[1110],{"type":49,"value":837},{"type":43,"tag":83,"props":1112,"children":1113},{"style":102},[1114],{"type":49,"value":842},{"type":43,"tag":83,"props":1116,"children":1118},{"style":1117},"--shiki-light:#F76D47;--shiki-default:#F78C6C;--shiki-dark:#F78C6C",[1119],{"type":49,"value":1120}," 27",{"type":43,"tag":83,"props":1122,"children":1123},{"style":102},[1124],{"type":49,"value":861},{"type":43,"tag":83,"props":1126,"children":1127},{"style":154},[1128],{"type":49,"value":1129},"                        \u002F\u002F from the API\n",{"type":43,"tag":83,"props":1131,"children":1133},{"class":85,"line":1132},10,[1134,1138,1143,1147,1151,1155,1160,1164],{"type":43,"tag":83,"props":1135,"children":1136},{"style":102},[1137],{"type":49,"value":938},{"type":43,"tag":83,"props":1139,"children":1140},{"style":108},[1141],{"type":49,"value":1142},"hardware",{"type":43,"tag":83,"props":1144,"children":1145},{"style":102},[1146],{"type":49,"value":837},{"type":43,"tag":83,"props":1148,"children":1149},{"style":102},[1150],{"type":49,"value":842},{"type":43,"tag":83,"props":1152,"children":1153},{"style":102},[1154],{"type":49,"value":847},{"type":43,"tag":83,"props":1156,"children":1157},{"style":96},[1158],{"type":49,"value":1159},"zero-a10g",{"type":43,"tag":83,"props":1161,"children":1162},{"style":102},[1163],{"type":49,"value":837},{"type":43,"tag":83,"props":1165,"children":1166},{"style":154},[1167],{"type":49,"value":1168},"             \u002F\u002F runtime.hardware.current\n",{"type":43,"tag":83,"props":1170,"children":1172},{"class":85,"line":1171},11,[1173],{"type":43,"tag":83,"props":1174,"children":1175},{"style":102},[1176],{"type":49,"value":1177},"  },\n",{"type":43,"tag":83,"props":1179,"children":1181},{"class":85,"line":1180},12,[1182,1186,1190,1194,1198,1202,1207,1211,1215],{"type":43,"tag":83,"props":1183,"children":1184},{"style":102},[1185],{"type":49,"value":826},{"type":43,"tag":83,"props":1187,"children":1188},{"style":829},[1189],{"type":49,"value":204},{"type":43,"tag":83,"props":1191,"children":1192},{"style":102},[1193],{"type":49,"value":837},{"type":43,"tag":83,"props":1195,"children":1196},{"style":102},[1197],{"type":49,"value":842},{"type":43,"tag":83,"props":1199,"children":1200},{"style":102},[1201],{"type":49,"value":847},{"type":43,"tag":83,"props":1203,"children":1204},{"style":96},[1205],{"type":49,"value":1206},"live output · zero-a10g",{"type":43,"tag":83,"props":1208,"children":1209},{"style":102},[1210],{"type":49,"value":837},{"type":43,"tag":83,"props":1212,"children":1213},{"style":102},[1214],{"type":49,"value":861},{"type":43,"tag":83,"props":1216,"children":1217},{"style":154},[1218],{"type":49,"value":1219},"    \u002F\u002F small honest footer note; use \"mock demo assets\" for mocks\n",{"type":43,"tag":83,"props":1221,"children":1223},{"class":85,"line":1222},13,[1224,1228,1233,1237,1241,1245,1250,1254,1258],{"type":43,"tag":83,"props":1225,"children":1226},{"style":102},[1227],{"type":49,"value":826},{"type":43,"tag":83,"props":1229,"children":1230},{"style":829},[1231],{"type":49,"value":1232},"button",{"type":43,"tag":83,"props":1234,"children":1235},{"style":102},[1236],{"type":49,"value":837},{"type":43,"tag":83,"props":1238,"children":1239},{"style":102},[1240],{"type":49,"value":842},{"type":43,"tag":83,"props":1242,"children":1243},{"style":102},[1244],{"type":49,"value":847},{"type":43,"tag":83,"props":1246,"children":1247},{"style":96},[1248],{"type":49,"value":1249},"Generate",{"type":43,"tag":83,"props":1251,"children":1252},{"style":102},[1253],{"type":49,"value":837},{"type":43,"tag":83,"props":1255,"children":1256},{"style":102},[1257],{"type":49,"value":861},{"type":43,"tag":83,"props":1259,"children":1260},{"style":154},[1261],{"type":49,"value":1262},"                 \u002F\u002F optional; e.g. \"Enhance\", \"Animate\"\n",{"type":43,"tag":83,"props":1264,"children":1266},{"class":85,"line":1265},14,[1267,1271,1276,1280,1284,1288,1293,1297,1301],{"type":43,"tag":83,"props":1268,"children":1269},{"style":102},[1270],{"type":49,"value":826},{"type":43,"tag":83,"props":1272,"children":1273},{"style":829},[1274],{"type":49,"value":1275},"input_label",{"type":43,"tag":83,"props":1277,"children":1278},{"style":102},[1279],{"type":49,"value":837},{"type":43,"tag":83,"props":1281,"children":1282},{"style":102},[1283],{"type":49,"value":842},{"type":43,"tag":83,"props":1285,"children":1286},{"style":102},[1287],{"type":49,"value":847},{"type":43,"tag":83,"props":1289,"children":1290},{"style":96},[1291],{"type":49,"value":1292},"source image",{"type":43,"tag":83,"props":1294,"children":1295},{"style":102},[1296],{"type":49,"value":837},{"type":43,"tag":83,"props":1298,"children":1299},{"style":102},[1300],{"type":49,"value":861},{"type":43,"tag":83,"props":1302,"children":1303},{"style":154},[1304],{"type":49,"value":1305},"        \u002F\u002F optional label overrides\n",{"type":43,"tag":83,"props":1307,"children":1309},{"class":85,"line":1308},15,[1310,1314,1319,1323,1327,1331,1336,1340],{"type":43,"tag":83,"props":1311,"children":1312},{"style":102},[1313],{"type":49,"value":826},{"type":43,"tag":83,"props":1315,"children":1316},{"style":829},[1317],{"type":49,"value":1318},"prompt_label",{"type":43,"tag":83,"props":1320,"children":1321},{"style":102},[1322],{"type":49,"value":837},{"type":43,"tag":83,"props":1324,"children":1325},{"style":102},[1326],{"type":49,"value":842},{"type":43,"tag":83,"props":1328,"children":1329},{"style":102},[1330],{"type":49,"value":847},{"type":43,"tag":83,"props":1332,"children":1333},{"style":96},[1334],{"type":49,"value":1335},"edit instruction",{"type":43,"tag":83,"props":1337,"children":1338},{"style":102},[1339],{"type":49,"value":837},{"type":43,"tag":83,"props":1341,"children":1342},{"style":102},[1343],{"type":49,"value":903},{"type":43,"tag":83,"props":1345,"children":1347},{"class":85,"line":1346},16,[1348,1352,1357,1361,1365,1369,1374,1378],{"type":43,"tag":83,"props":1349,"children":1350},{"style":102},[1351],{"type":49,"value":826},{"type":43,"tag":83,"props":1353,"children":1354},{"style":829},[1355],{"type":49,"value":1356},"output_label",{"type":43,"tag":83,"props":1358,"children":1359},{"style":102},[1360],{"type":49,"value":837},{"type":43,"tag":83,"props":1362,"children":1363},{"style":102},[1364],{"type":49,"value":842},{"type":43,"tag":83,"props":1366,"children":1367},{"style":102},[1368],{"type":49,"value":847},{"type":43,"tag":83,"props":1370,"children":1371},{"style":96},[1372],{"type":49,"value":1373},"edited image",{"type":43,"tag":83,"props":1375,"children":1376},{"style":102},[1377],{"type":49,"value":837},{"type":43,"tag":83,"props":1379,"children":1380},{"style":102},[1381],{"type":49,"value":903},{"type":43,"tag":83,"props":1383,"children":1385},{"class":85,"line":1384},17,[1386,1390,1395,1399,1403,1408,1412],{"type":43,"tag":83,"props":1387,"children":1388},{"style":102},[1389],{"type":49,"value":826},{"type":43,"tag":83,"props":1391,"children":1392},{"style":829},[1393],{"type":49,"value":1394},"audio_gain",{"type":43,"tag":83,"props":1396,"children":1397},{"style":102},[1398],{"type":49,"value":837},{"type":43,"tag":83,"props":1400,"children":1401},{"style":102},[1402],{"type":49,"value":842},{"type":43,"tag":83,"props":1404,"children":1405},{"style":1117},[1406],{"type":49,"value":1407}," 2.0",{"type":43,"tag":83,"props":1409,"children":1410},{"style":102},[1411],{"type":49,"value":861},{"type":43,"tag":83,"props":1413,"children":1414},{"style":154},[1415],{"type":49,"value":1416},"                    \u002F\u002F optional; default 2.0 (with limiter)\n",{"type":43,"tag":83,"props":1418,"children":1420},{"class":85,"line":1419},18,[1421,1425,1430,1434,1438,1443],{"type":43,"tag":83,"props":1422,"children":1423},{"style":102},[1424],{"type":49,"value":826},{"type":43,"tag":83,"props":1426,"children":1427},{"style":829},[1428],{"type":49,"value":1429},"hero",{"type":43,"tag":83,"props":1431,"children":1432},{"style":102},[1433],{"type":49,"value":837},{"type":43,"tag":83,"props":1435,"children":1436},{"style":102},[1437],{"type":49,"value":842},{"type":43,"tag":83,"props":1439,"children":1440},{"style":102},[1441],{"type":49,"value":1442}," false",{"type":43,"tag":83,"props":1444,"children":1445},{"style":154},[1446],{"type":49,"value":1447},"                         \u002F\u002F optional; set false to skip the full-frame intro\n",{"type":43,"tag":83,"props":1449,"children":1451},{"class":85,"line":1450},19,[1452],{"type":43,"tag":83,"props":1453,"children":1454},{"style":102},[1455],{"type":49,"value":1456},"}\n",{"type":43,"tag":52,"props":1458,"children":1459},{},[1460,1462,1468],{"type":49,"value":1461},"Plus the template-specific fields from the table. All asset paths are absolute paths on\nthis machine. Store working assets under ",{"type":43,"tag":63,"props":1463,"children":1465},{"className":1464},[],[1466],{"type":49,"value":1467},"\u002Froot\u002Fcode\u002F\u003Cproject-dir>\u002F",{"type":49,"value":1469}," as usual.",{"type":43,"tag":159,"props":1471,"children":1473},{"id":1472},"workflow",[1474],{"type":49,"value":1475},"Workflow",{"type":43,"tag":166,"props":1477,"children":1478},{},[1479,1498,1503,1529,1542,1547],{"type":43,"tag":170,"props":1480,"children":1481},{},[1482,1488,1490,1496],{"type":43,"tag":63,"props":1483,"children":1485},{"className":1484},[],[1486],{"type":49,"value":1487},"curl -s https:\u002F\u002Fhuggingface.co\u002Fapi\u002Fspaces\u002F\u003Cid>",{"type":49,"value":1489}," → metadata for ",{"type":43,"tag":63,"props":1491,"children":1493},{"className":1492},[],[1494],{"type":49,"value":1495},"space{}",{"type":49,"value":1497}," fields.",{"type":43,"tag":170,"props":1499,"children":1500},{},[1501],{"type":49,"value":1502},"Pick the template from the table. If the Space's task fits none of them, say so and pick\nthe closest; do NOT invent a new design.",{"type":43,"tag":170,"props":1504,"children":1505},{},[1506,1508,1513,1515,1520,1522,1527],{"type":49,"value":1507},"Capture assets: choose a good input (the Space's own example assets are ideal), run the\nlive Space with ",{"type":43,"tag":63,"props":1509,"children":1511},{"className":1510},[],[1512],{"type":49,"value":196},{"type":49,"value":1514},", download the real output. Verify the output file with\n",{"type":43,"tag":63,"props":1516,"children":1518},{"className":1517},[],[1519],{"type":49,"value":320},{"type":49,"value":1521},"\u002F",{"type":43,"tag":63,"props":1523,"children":1525},{"className":1524},[],[1526],{"type":49,"value":374},{"type":49,"value":1528}," before rendering.",{"type":43,"tag":170,"props":1530,"children":1531},{},[1532,1534,1540],{"type":49,"value":1533},"Write ",{"type":43,"tag":63,"props":1535,"children":1537},{"className":1536},[],[1538],{"type":49,"value":1539},"manifests\u002F\u003Cname>.json",{"type":49,"value":1541}," in the kit.",{"type":43,"tag":170,"props":1543,"children":1544},{},[1545],{"type":49,"value":1546},"Render (command above). First run downloads fonts\u002Femoji (needs network) and, for video\ninputs, extracts frames. A render takes roughly 3–5 minutes — do not assume it hung.",{"type":43,"tag":170,"props":1548,"children":1549},{},[1550],{"type":49,"value":1551},"Verify MP4 + contact sheet, then deliver the MP4 to the user as a media message.",{"type":43,"tag":159,"props":1553,"children":1555},{"id":1554},"codex-dispatch-openclaw-mandatory",[1556],{"type":49,"value":1557},"Codex dispatch (OpenClaw — mandatory)",{"type":43,"tag":52,"props":1559,"children":1560},{},[1561,1563,1569],{"type":49,"value":1562},"Demo-video requests are heavy (live inference + a ~2 min render). Do NOT run inference or\nrendering inline in the main session — it blocks you for many minutes. ALWAYS dispatch a\ncodex session per the codex-monitor skill (setsid, ",{"type":43,"tag":63,"props":1564,"children":1566},{"className":1565},[],[1567],{"type":49,"value":1568},"\u003C\u002Fdev\u002Fnull",{"type":49,"value":1570},", CODEX_HOME pinned) and let\nthe CODEX_FINISHED wake drive verification + delivery. Include in the codex prompt:\n\"Read \u002Froot\u002Fcode\u002Fspace-demo-kit\u002Fskills\u002Fspace-demo-video\u002FSKILL.md and follow it exactly.\nUse the kit's render.py for ALL visuals; do not design any frame yourself.\"",{"type":43,"tag":159,"props":1572,"children":1574},{"id":1573},"gotchas",[1575],{"type":49,"value":1576},"Gotchas",{"type":43,"tag":356,"props":1578,"children":1579},{},[1580,1585,1590,1608,1618,1628],{"type":43,"tag":170,"props":1581,"children":1582},{},[1583],{"type":49,"value":1584},"Long Space ids auto-shrink in the footer; long repo names get ellipsized in the header — fine.",{"type":43,"tag":170,"props":1586,"children":1587},{},[1588],{"type":49,"value":1589},"The renderer needs Playwright + Chromium (already installed system-wide here).",{"type":43,"tag":170,"props":1591,"children":1592},{},[1593,1595,1600,1601,1606],{"type":49,"value":1594},"Audio templates (",{"type":43,"tag":63,"props":1596,"children":1598},{"className":1597},[],[1599],{"type":49,"value":601},{"type":49,"value":232},{"type":43,"tag":63,"props":1602,"children":1604},{"className":1603},[],[1605],{"type":49,"value":423},{"type":49,"value":1607},") mux the real audio automatically, delayed to the moment\nplayback starts on screen, at 2x gain with a limiter.",{"type":43,"tag":170,"props":1609,"children":1610},{},[1611,1616],{"type":43,"tag":63,"props":1612,"children":1614},{"className":1613},[],[1615],{"type":49,"value":673},{"type":49,"value":1617}," wants exactly 4 outputs for the 2x2 grid.",{"type":43,"tag":170,"props":1619,"children":1620},{},[1621,1626],{"type":43,"tag":63,"props":1622,"children":1624},{"className":1623},[],[1625],{"type":49,"value":721},{"type":49,"value":1627}," answers auto-fit the output panel: keep them 30–60 words so the type stays large.",{"type":43,"tag":170,"props":1629,"children":1630},{},[1631,1633,1639,1641,1647],{"type":49,"value":1632},"Do not edit ",{"type":43,"tag":63,"props":1634,"children":1636},{"className":1635},[],[1637],{"type":49,"value":1638},"templates\u002Fapp-window.html",{"type":49,"value":1640}," or ",{"type":43,"tag":63,"props":1642,"children":1644},{"className":1643},[],[1645],{"type":49,"value":1646},"render.py",{"type":49,"value":1648}," for a one-off video. If a Space\ngenuinely needs a new mode, report that instead of hacking around the kit.",{"type":43,"tag":1650,"props":1651,"children":1652},"style",{},[1653],{"type":49,"value":1654},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1656,"total":86},[1657],{"slug":4,"name":4,"fn":5,"description":6,"org":1658,"tags":1659,"stars":27,"repoUrl":28,"updatedAt":29},{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1660,1661,1662,1663,1664],{"name":19,"slug":20,"type":15},{"name":9,"slug":17,"type":15},{"name":22,"slug":23,"type":15},{"name":13,"slug":14,"type":15},{"name":25,"slug":26,"type":15},{"items":1666,"total":1843},[1667,1689,1705,1722,1736,1749,1762,1777,1791,1801,1814,1828],{"slug":1668,"name":1668,"fn":1669,"description":1670,"org":1671,"tags":1672,"stars":1686,"repoUrl":1687,"updatedAt":1688},"train-sentence-transformers","train sentence-transformers models","Train or fine-tune sentence-transformers models across `SentenceTransformer` (bi-encoder; dense or static embedding model; for retrieval, similarity, clustering, classification, paraphrase mining, dedup, multimodal), `CrossEncoder` (reranker; pair scoring for two-stage retrieval \u002F pair classification), and `SparseEncoder` (SPLADE, sparse embedding model; for learned-sparse retrieval). Covers loss selection, hard-negative mining, evaluators, distillation, LoRA, Matryoshka, and Hugging Face Hub publishing. Use for any sentence-transformers training task.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1673,1676,1677,1680,1683],{"name":1674,"slug":1675,"type":15},"Deep Learning","deep-learning",{"name":9,"slug":17,"type":15},{"name":1678,"slug":1679,"type":15},"LLM","llm",{"name":1681,"slug":1682,"type":15},"Python","python",{"name":1684,"slug":1685,"type":15},"Search","search",18914,"https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fsentence-transformers","2026-05-08T05:09:16.820066",{"slug":1690,"name":1690,"fn":1691,"description":1692,"org":1693,"tags":1694,"stars":1702,"repoUrl":1703,"updatedAt":1704},"trl-training","train and fine-tune LLMs with TRL","Train and fine-tune transformer language models using TRL (Transformers Reinforcement Learning). Supports SFT, DPO, GRPO, KTO, RLOO and Reward Model training via CLI commands.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1695,1698,1699,1700,1701],{"name":1696,"slug":1697,"type":15},"AI Infrastructure","ai-infrastructure",{"name":1674,"slug":1675,"type":15},{"name":9,"slug":17,"type":15},{"name":1678,"slug":1679,"type":15},{"name":1681,"slug":1682,"type":15},18850,"https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Ftrl","2026-04-06T18:25:32.746828",{"slug":1706,"name":1706,"fn":1707,"description":1708,"org":1709,"tags":1710,"stars":1719,"repoUrl":1720,"updatedAt":1721},"hf-cli","manage Hugging Face Hub resources via CLI","Hugging Face Hub CLI (`hf`) for downloading, uploading, and managing models, datasets, spaces, buckets, repos, papers, jobs, and more on the Hugging Face Hub. Use when: handling authentication; managing local cache; managing Hugging Face Buckets; running or scheduling jobs on Hugging Face infrastructure; managing Hugging Face repos; discussions and pull requests; browsing models, datasets and spaces; reading, searching, or browsing academic papers; managing collections; querying datasets; configuring spaces; setting up webhooks; or deploying and managing HF Inference Endpoints. Make sure to use this skill whenever the user mentions 'hf', 'huggingface', 'Hugging Face', 'huggingface-cli', or 'hugging face cli', or wants to do anything related to the Hugging Face ecosystem and to AI and ML in general. Also use for cloud storage needs like training checkpoints, data pipelines, or agent traces. Use even if the user doesn't explicitly ask for a CLI command. Replaces the deprecated `huggingface-cli`.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1711,1714,1717,1718],{"name":1712,"slug":1713,"type":15},"CLI","cli",{"name":1715,"slug":1716,"type":15},"Datasets","datasets",{"name":9,"slug":17,"type":15},{"name":1678,"slug":1679,"type":15},10861,"https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fskills","2026-04-06T18:25:34.020855",{"slug":1723,"name":1723,"fn":1724,"description":1725,"org":1726,"tags":1727,"stars":1719,"repoUrl":1720,"updatedAt":1735},"hf-cloud-aws-context-discovery","discover local AWS environment context","Discover the user's local AWS context (active profile, region, account ID, caller identity) at the start of any AWS task. Use this skill before any other AWS work — deploying to SageMaker, creating resources, calling AWS APIs, or anything that touches an AWS account. Use it especially when the user has not specified a region or profile explicitly, when they say things like \"use my AWS account\", \"deploy to AWS\", \"use my profile\", or when about to make any AWS CLI or SDK call. Never guess the region or account ID — always use this skill to read it from the local configuration first.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1728,1731,1732],{"name":1729,"slug":1730,"type":15},"AWS","aws",{"name":1712,"slug":1713,"type":15},{"name":1733,"slug":1734,"type":15},"Configuration","configuration","2026-07-08T05:55:33.716099",{"slug":1737,"name":1737,"fn":1738,"description":1739,"org":1740,"tags":1741,"stars":1719,"repoUrl":1720,"updatedAt":1748},"hf-cloud-python-env-setup","set up Python environments for AWS","Set up an isolated Python environment for SageMaker \u002F AWS work, with the right Python version and current boto3. Use this skill whenever Python code will be executed for a SageMaker deployment, training job, or any AWS automation — including when about to run `pip install`, when about to invoke `boto3`, when creating or activating a virtualenv, or when the user asks to \"set up the environment\". Never use system Python and never `pip install` into it. Always isolate. This skill prevents the most common failure modes: wrong Python version, dependency conflicts, and stale SDKs.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1742,1743,1744,1747],{"name":1696,"slug":1697,"type":15},{"name":1729,"slug":1730,"type":15},{"name":1745,"slug":1746,"type":15},"Engineering","engineering",{"name":1681,"slug":1682,"type":15},"2026-07-08T05:55:32.505017",{"slug":1750,"name":1750,"fn":1751,"description":1752,"org":1753,"tags":1754,"stars":1719,"repoUrl":1720,"updatedAt":1761},"hf-cloud-sagemaker-deployment-planner","plan model deployments to Amazon SageMaker","Plan and coordinate the deployment of a model to Amazon SageMaker AI. Use this skill whenever the user wants to deploy, host, serve, or expose a model on SageMaker or AWS — including phrases like \"deploy a model\", \"host this LLM on AWS\", \"serve this embedding model\", \"deploy a reranker\", \"deploy a text-to-image \u002F diffusion model\", \"host this for async inference\", \"create an endpoint\", \"serve my fine-tuned model\", or any request that involves making a model available for inference on AWS. Use this even when the user is vague (e.g. \"I just want to get this running on AWS, you figure it out\"). Works for text-generation LLMs, embedding models, rerankers, classifiers, text-to-image \u002F diffusion models — picks the right serving stack and chooses between real-time and async inference. This is the entry-point skill for SageMaker deployment work — it asks clarifying questions, picks a deployment pathway, and coordinates the other deployment skills.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1755,1756,1757,1760],{"name":1696,"slug":1697,"type":15},{"name":1729,"slug":1730,"type":15},{"name":1758,"slug":1759,"type":15},"Deployment","deployment",{"name":9,"slug":17,"type":15},"2026-07-08T05:55:37.387689",{"slug":1763,"name":1763,"fn":1764,"description":1765,"org":1766,"tags":1767,"stars":1719,"repoUrl":1720,"updatedAt":1776},"hf-cloud-sagemaker-iam-preflight","configure SageMaker IAM roles","Ensure a usable SageMaker execution role exists before deploying or training. Use this skill whenever about to create a SageMaker endpoint, model, training job, or any resource that requires an execution role. Use it especially when the user has not provided a role ARN explicitly, when scripts are about to call `iam:CreateRole`, or when an AccessDenied error mentions an IAM action. Never blindly call `iam:CreateRole` — always check for existing roles first. This skill prevents the most common SageMaker deployment failure: trying to create IAM resources from an SSO principal that has no IAM write permissions.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1768,1769,1770,1773],{"name":1696,"slug":1697,"type":15},{"name":1729,"slug":1730,"type":15},{"name":1771,"slug":1772,"type":15},"Permissions","permissions",{"name":1774,"slug":1775,"type":15},"Security","security","2026-07-08T05:55:34.948771",{"slug":1778,"name":1778,"fn":1779,"description":1780,"org":1781,"tags":1782,"stars":1719,"repoUrl":1720,"updatedAt":1790},"hf-cloud-sagemaker-production-defaults","create production-ready SageMaker endpoints","Create a SageMaker endpoint (real-time or async) with autoscaling, CloudWatch alarms, and tagging enabled by default. Use this skill whenever about to create a SageMaker endpoint, write deployment code that calls `create_endpoint`, or finalize a deployment after the image URI and IAM role are known. Provides deploy.py for real-time endpoints and deploy_async.py for async endpoints (with genuine scale-to-zero support). This is the last step in the SageMaker deployment workflow. Never generate a bare `create_endpoint` call without these defaults — endpoints without autoscaling or alarms are demos, not deployments.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1783,1784,1785,1786,1787],{"name":1696,"slug":1697,"type":15},{"name":1729,"slug":1730,"type":15},{"name":1758,"slug":1759,"type":15},{"name":9,"slug":17,"type":15},{"name":1788,"slug":1789,"type":15},"Monitoring","monitoring","2026-07-08T05:55:38.664702",{"slug":1792,"name":1792,"fn":1793,"description":1794,"org":1795,"tags":1796,"stars":1719,"repoUrl":1720,"updatedAt":1800},"hf-cloud-serving-image-selection","select SageMaker serving containers","Pick the right serving container for a SageMaker model deployment and find its current image URI. Use this skill whenever about to deploy a model to a SageMaker endpoint and an image URI needs to be chosen — including when the user says \"deploy this LLM\", \"host this HuggingFace model\", \"serve this fine-tuned model\", \"deploy this embedding model\", \"host a reranker\", \"serve a sentence-transformers model\", or when about to hardcode any container URI in deployment code. HuggingFace-curated Deep Learning Containers are ALWAYS preferred: HuggingFace vLLM (LLMs and generative rerankers), HuggingFace vLLM-Omni (multimodal), TEI (embeddings\u002Fcross-encoder rerankers), HF Inference Toolkit (other transformers). Generic images (AWS vLLM, DJL-LMI, SGLang) are used only when no HuggingFace image is compatible — never merely because they carry a newer version. Never hardcode a container URI from memory and never default to TGI. Prevents stale-image failures and wrong-region URIs.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1797,1798,1799],{"name":1696,"slug":1697,"type":15},{"name":1729,"slug":1730,"type":15},{"name":1758,"slug":1759,"type":15},"2026-07-08T05:55:36.173465",{"slug":1802,"name":1802,"fn":1803,"description":1804,"org":1805,"tags":1806,"stars":1719,"repoUrl":1720,"updatedAt":1813},"hf-mcp","access Hugging Face Hub via MCP","Use Hugging Face Hub via MCP server tools. Search models, datasets, Spaces, papers. Get repo details, fetch documentation, run compute jobs, and use Gradio Spaces as AI tools. Available when connected to the HF MCP server.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1807,1808,1809,1810],{"name":1715,"slug":1716,"type":15},{"name":9,"slug":17,"type":15},{"name":1678,"slug":1679,"type":15},{"name":1811,"slug":1812,"type":15},"MCP","mcp","2026-04-06T18:25:50.364185",{"slug":1815,"name":1815,"fn":1816,"description":1817,"org":1818,"tags":1819,"stars":1719,"repoUrl":1720,"updatedAt":1827},"hf-mem","estimate memory for Hugging Face models","Hugging Face CLI to estimate the required memory to load Safetensors or GGUF model weights for inference from the Hugging Face Hub",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1820,1821,1822,1823,1824],{"name":1696,"slug":1697,"type":15},{"name":1712,"slug":1713,"type":15},{"name":9,"slug":17,"type":15},{"name":1678,"slug":1679,"type":15},{"name":1825,"slug":1826,"type":15},"Performance","performance","2026-06-13T07:23:57.101435",{"slug":1829,"name":1829,"fn":1830,"description":1831,"org":1832,"tags":1833,"stars":1719,"repoUrl":1720,"updatedAt":1842},"huggingface-best","find and compare Hugging Face models","Use when the user asks about finding the best, top, or recommended model for a task, wants to know what AI model to use, or wants to compare models by benchmark scores. Triggers on: \"best model for X\", \"what model should I use for\", \"top models for [task]\", \"which model runs on my laptop\u002Fmachine\u002Fdevice\", \"recommend a model for\", \"what LLM should I use for\", \"compare models for\", \"what's state of the art for\", or any question about choosing an AI model for a specific use case. Always use this skill when the user wants model recommendations or comparisons, even if they don't explicitly mention HuggingFace or benchmarks.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":8},[1834,1837,1838,1839],{"name":1835,"slug":1836,"type":15},"Analytics","analytics",{"name":9,"slug":17,"type":15},{"name":1678,"slug":1679,"type":15},{"name":1840,"slug":1841,"type":15},"Research","research","2026-04-24T05:09:45.870658",37]