{"schema_version":"1","site":"https://decosa.ai","id":"studio","num":"04","name":"Decosa Studio","tool_name":"Make a story, a song or a music video","short":"Studio","blurb":"Make things with your people in them: a picture book read in your own voice, a song, a music video, an animatic or a dub. Add your cast once and every pack uses it.","status":"live","labels":{"industry":["creative-media","personal-family"],"job":["generate"],"input":["text"],"deploy":["hosted","selfhost"],"status":"live","output":["media"],"data":["none"],"hardware":"gpu-96","licence":"community"},"industries":["creative-media","personal-family"],"runs_in":["hosted","selfhost"],"part_of":[],"built_from":["studio-render"],"models":"Qwen3.8-27B · FLUX.2 [klein] 4B · Chatterbox · ACE-Step · MiniMax-Music3 · MiniMax H3","where":"Hosted or self-host","hardware":"1× RTX PRO 6000 (96 GB)","final_artifact":"Rendered audio, images and video.","self_host_first":false,"verification":{"receipt_coverage":"partial","summary":"Partial (seeded re-render)","manual_qa":{"hosted":{"date":"2026-09-25","result":"partial","p50_ms":null,"p95_ms":null,"runs":null,"receipts_per_run":1,"cost_per_run_usd":null},"selfhost":{"date":"2026-09-25","result":"pass","method":"Fresh clone of decosa-api; the api image plus ffmpeg running the repo's own studio worker against an already-running ComfyUI on the same box (no render or ComfyUI containers built, no new model loads).","notes":"Verified on 2026-09-25: the image builds, the service starts, and the prompt's MiniMax-Music3 smoke job (seed 5501) finished in 40 s against a local ComfyUI equivalent to the documented one; model-server startup and the image and video paths were not re-verified. Two runs of the same seed gave bit-identical decoded PCM. An api restart mid-render marked that job failed with a reason and ran the queued one (72 s). Fixed on the way: the api image left out services/ (every render failed), jobs were lost on restart."},"known_limits":["Renders share one GPU with the live audio demos: a job pauses while a live session runs.","Each token or key can queue 3 jobs; hosted video needs an API key and fal credit.","Receipts for renders are our signed statement of model, seed, weights and output hash; nobody re-renders to check them.","Wan2.1 video takes about 35 minutes per 5 s clip on the shared GPU."],"nightly_covers":null},"nightly":"https://api.decosa.ai/verify/status"},"eval_summary":{"metrics":[],"dataset":"No quality eval. One reproducibility check on our server: same-seed re-render is bit-identical for MiniMax-Music3 and Qwen-Image-2512, not identical for ACE-Step.","held_out":false,"caveats":["Output quality not measured yet on any tier.","The same-seed check shows reproducibility, not quality."],"date":"2026-09-23","doc_url":null},"stack":{"summary":"A render queue for songs, images and short videos, running open-weight models that load onto one GPU one at a time. Private checkpoints and LoRAs stay on your disk. Voice with consent-gated cloning is planned. It is built for creative teams and indie studios that want predictable costs and control over their weights.","tagline":"Music, images and video from open weights on your own GPU, with a render queue and a pre-rendered gallery.","deployment":"hosted-or-self-host","regulatory_note":"Hosted demo models: MiniMax-Music3 (show \"MiniMax-Music3\" prominently; abuse and IP safeguards; authorization above USD 20M revenue), ACE-Step (MIT), Qwen-Image-2512 (Apache-2.0) and Wan2.1-T2V-14B (Apache-2.0). LTX-2.5 and MiniMax-H3 are self-host only because of their licences. Voice cloning needs documented consent from the speaker; never clone a real person's voice or likeness. Excluded: Qwen-Image-2.1 (research-only licence) and OmniVoice (CC-BY-NC weights).","components":[{"id":"music","role":"Music generation, fast drafts (text and lyrics to song)","name":"ACE-Step 1.5 turbo + 5Hz LM 1.7B","hf_repo":"ACE-Step/Ace-Step1.5","license":"MIT","params":"2.39B DiT + 1.85B LM","quant":"BF16","vram_gb":14.6,"memory_gb_estimate":null,"engine":"ACE-Step 1.5 Python inference (acestep.inference.generate_music, LM backend pt)","receipt_coverage":"partial","in_hosted_demo":true,"tiers":["lite"],"alternative_to":null},{"id":"music3","role":"Music generation (songs with vocals and lyrics)","name":"MiniMax-Music3","hf_repo":"MiniMaxAI/MiniMax-Music3","license":"MiniMax-Music3 Community License","params":"8.58B LM + 2.43B flow transformer","quant":"FP16 DiT + BF16 pruned text encoder (Comfy-Org repackage)","vram_gb":null,"memory_gb_estimate":null,"engine":"ComfyUI core MiniMax Music 3 nodes (MiniMaxMusic3TextEncode, KSampler euler/simple, 30 steps, cfg 1.7)","receipt_coverage":"partial","in_hosted_demo":true,"tiers":[],"alternative_to":null},{"id":"image-lite","role":"Image generation, fast drafts","name":"Z-Image-Turbo","hf_repo":"Tongyi-MAI/Z-Image-Turbo","license":"Apache-2.0","params":null,"quant":"BF16","vram_gb":null,"memory_gb_estimate":null,"engine":"ComfyUI core (Z-Image nodes)","receipt_coverage":"partial","in_hosted_demo":false,"tiers":["lite"],"alternative_to":null},{"id":"image","role":"Image generation","name":"Qwen-Image-2512","hf_repo":"Qwen/Qwen-Image-2512","license":"Apache-2.0","params":"20.4B (transformer)","quant":"BF16","vram_gb":41.6,"memory_gb_estimate":null,"engine":"diffusers QwenImagePipeline with model CPU offload","receipt_coverage":"partial","in_hosted_demo":true,"tiers":[],"alternative_to":null},{"id":"video-lite","role":"Video generation, drafts","name":"Wan2.1-T2V-1.3B","hf_repo":"Wan-AI/Wan2.1-T2V-1.3B","license":"Apache-2.0","params":"1.3B","quant":"FP32 weights (original release)","vram_gb":null,"memory_gb_estimate":null,"engine":"ComfyUI core Wan nodes","receipt_coverage":"partial","in_hosted_demo":false,"tiers":["lite"],"alternative_to":null},{"id":"video-h3-fal","role":"Hosted video (default): 5 s clips with audio","name":"MiniMax H3 Max (via fal)","hf_repo":null,"license":"H3's own licence excludes the US; used here only through fal's hosted endpoints, listed as commercial use under fal's MiniMax partnership","params":null,"quant":null,"vram_gb":null,"memory_gb_estimate":null,"engine":"fal queue API, minimax/h3-max/text-to-video (1080P, 16:9, 5 s)","receipt_coverage":"none","in_hosted_demo":true,"tiers":["standard"],"alternative_to":null},{"id":"video","role":"Video generation (text to video, no audio)","name":"Wan2.1-T2V-14B","hf_repo":"Wan-AI/Wan2.1-T2V-14B","license":"Apache-2.0","params":"14B","quant":"FP32 weights (original release), run by ComfyUI","vram_gb":null,"memory_gb_estimate":null,"engine":"ComfyUI core Wan nodes (UNETLoader, umt5-xxl text encoder, Wan2.1 VAE, KSampler uni_pc/simple, 50 steps, cfg 5.0, shift 5.0)","receipt_coverage":"partial","in_hosted_demo":false,"tiers":["lite"],"alternative_to":null},{"id":"video-ltx","role":"Video with synchronized audio, self-host only","name":"LTX-2.5 22B distilled","hf_repo":"Lightricks/LTX-2.5","license":"LTX-2 Community License","params":"21.0B (transformer)","quant":"INT8 (comfy int8-convrot transformer and Gemma text encoder)","vram_gb":null,"memory_gb_estimate":null,"engine":"ComfyUI core LTX nodes (two-stage distilled: sample, latent 2x upscale, refine)","receipt_coverage":"partial","in_hosted_demo":false,"tiers":["best"],"alternative_to":null},{"id":"video-h3","role":"Video with synchronized audio and reference control, self-host only where licensed","name":"MiniMax-H3 (licence pending)","hf_repo":"MiniMaxAI/MiniMax-H3","license":"MiniMax H3 Community License today (it excludes the US, EU, UK and Korea); a licence for our use is pending","params":"33.1B (transformer) + 33.4B (text encoder)","quant":"BF16 with CPU offload","vram_gb":null,"memory_gb_estimate":133,"engine":"diffusers ModularPipeline (MiniMax-H3 blocks)","receipt_coverage":"partial","in_hosted_demo":false,"tiers":["best"],"alternative_to":null},{"id":"voice","role":"Text to speech (preset voices)","name":"Kokoro-82M","hf_repo":"hexgrad/Kokoro-82M","license":"Apache-2.0","params":"82M","quant":"FP32","vram_gb":null,"memory_gb_estimate":null,"engine":"kokoro (Python package)","receipt_coverage":"none","in_hosted_demo":false,"tiers":[],"alternative_to":null},{"id":"voice-clone","role":"Voice cloning (consent-gated)","name":"IndexTTS-2.5","hf_repo":"IndexTeam/IndexTTS-2.5","license":"bilibili Model Use License (commercial use allowed below 100M MAU and RMB 1B yearly revenue; may not be used to improve other models)","params":null,"quant":null,"vram_gb":null,"memory_gb_estimate":null,"engine":"index-tts (github.com/index-tts/index-tts)","receipt_coverage":"none","in_hosted_demo":false,"tiers":[],"alternative_to":null}],"tiers":[{"id":"lite","label":"Lite · drafts on one 24–32 GB card","summary":"Fast music drafts, draft images and small draft video on one consumer card. Quality is below the standard tier, and Z-Image-Turbo and Wan 1.3B are not tested yet.","components":["music","image-lite","video-lite","voice","video"],"hardware":"1× 24–32 GB card (not tested)","quality_evidence":[{"metric":"output quality","value":"not measured yet","source":null}],"latency_note":"measured: ACE-Step in seconds per clip on our server; Z-Image-Turbo and Wan2.1-1.3B not measured yet","in_hosted_demo":false,"receipt_coverage":"partial","receipt_note":null,"hosting":null},{"id":"standard","label":"Standard · the hosted demo, about 50 GB of one 96 GB card","summary":"The hosted gallery and queue: MiniMax-Music3 songs, Qwen-Image-2512 images and Wan2.1-T2V-14B video (silent), with ACE-Step clips as well. Every model's licence allows hosted use. Hosted video runs on MiniMax H3 Max through fal at list price with no markup (API key, 2 per key per day, daily cap); Wan2.1 stays the fully open option.","components":["music3","music","image","video-h3-fal"],"hardware":"1× RTX PRO 6000 Blackwell 96 GB, about 50 GB free","quality_evidence":[{"metric":"output quality","value":"not measured yet","source":null},{"metric":"same-seed re-render","value":"bit-identical for MiniMax-Music3 and Qwen-Image-2512; not identical for ACE-Step","source":"measured on our server 2026-09-23"}],"latency_note":"measured on our server: Music3 about a minute per clip, an image in a couple of minutes, Wan 14B over half an hour per clip on shared memory","in_hosted_demo":true,"receipt_coverage":"partial","receipt_note":null,"hosting":null},{"id":"best","label":"Best · self-host only, a whole 96 GB card","summary":"Adds video with sound: MiniMax-H3 (licence pending; self-host available where its current licence covers you) or LTX-2.5 (below USD 10M revenue, not in a competing hosted service). Not in the hosted demo until the H3 licence lands.","components":["music3","image","video-ltx","video-h3"],"hardware":"1× RTX PRO 6000 Blackwell 96 GB, 256 GB RAM for MiniMax-H3","quality_evidence":[{"metric":"output quality","value":"not measured yet","source":null}],"latency_note":"measured: LTX-2.5 about a minute per clip with sound; MiniMax-H3 several minutes per clip","in_hosted_demo":false,"receipt_coverage":"partial","receipt_note":null,"hosting":null},{"id":"wanted","label":"Wanted · MiniMax H3 on two cards, no offload","summary":"The studio's best video model, faster. Both halves of MiniMax-H3 in BF16 on two 96 GB cards, so nothing is offloaded to RAM: faster renders (estimate, not measured). Self-host now where H3's current licence covers you; once the pending licence lands we add this compute for the hosted tier ourselves. Not a community-provider model.","components":["video-h3"],"hardware":"2x RTX PRO 6000 96 GB, no CPU offload (about 133 GB of BF16 weights; estimate)","quality_evidence":[{"metric":"render time per clip against one card with offload","value":"not measured yet","source":null}],"latency_note":"not measured yet","in_hosted_demo":false,"receipt_coverage":"none","receipt_note":"Self-host: your box signs the render record. Not a community-provider model.","hosting":"own-hardware"}],"alternates":[],"services":[{"name":"decosa-api","port":8445,"image":"${DECOSA_REGISTRY}/decosa-api:0.1.0","purpose":"Studio gallery (GET /studio/gallery), job queue (POST /studio/jobs, GET /studio/jobs/{id}) and static media under /studio/media; runs render jobs through DECOSA_STUDIO_WORKER=command."},{"name":"render","port":8460,"image":null,"purpose":"Local GPU render worker, built from source: loads one model at a time, renders, writes the file to a shared volume and returns its path."},{"name":"comfyui","port":8188,"image":null,"purpose":"MiniMax-Music3 and Wan2.1 graphs, reached only by the render worker; models are unloaded through POST /free after each job."}],"tools":[{"name":"ComfyUI","url":"https://github.com/comfyanonymous/ComfyUI","license":"GPL-3.0","purpose":"Runs the MiniMax-Music3 and Wan2.1 graphs (and LTX-2.5 for self-host); it is started only while a job needs it."},{"name":"diffusers","url":"https://github.com/huggingface/diffusers","license":"Apache-2.0","purpose":"Qwen-Image-2512 in the render worker (and MiniMax-H3 for self-host where licensed)."},{"name":"FFmpeg","url":"https://ffmpeg.org","license":"LGPL-2.1+","purpose":"Encodes WAV to MP3 and moves the MP4 index to the front for streaming."}],"hardware":[{"tier":"About 50 GB free on one RTX PRO 6000 Blackwell 96 GB, shared with other services","fits":true,"notes":"Measured 2026-09-23: the gallery's music, images and video rendered on GPU0 with 45–48 GB already in use by other services."},{"tier":"1× RTX PRO 6000 Blackwell 96 GB, 256 GB RAM (whole card)","fits":true,"notes":"Self-host best tier. MiniMax-H3 has run one job per card with about 115 GB offloaded to system RAM (measured before the licence review)."},{"tier":"1× 24–32 GB card","fits":null,"notes":"Not tested. ACE-Step (14.6 GB peak), Kokoro, Z-Image-Turbo and Wan2.1-1.3B are the realistic set; Qwen-Image-2512 peaked at 41.6 GB even with CPU offload."}],"latency":[{"lane":"music","typical_ms":3800,"source":"measured on our server 2026-09-23: ACE-Step 1.5 turbo, 30 s clip, 8 steps, mean of 4 renders (3.5–4.3 s), after a 16 s model load"},{"lane":"music","typical_ms":66000,"source":"measured on our server 2026-09-23: MiniMax-Music3 in ComfyUI, 45 s max duration, 30 steps, prompt execution including model load, 3 renders (43.0, 64.8, 72.6 s) on about 50 GB of free VRAM"},{"lane":"image","typical_ms":98750,"source":"measured on our server 2026-09-23: Qwen-Image-2512, 50 steps, 1328×1328 or 1664×928, model CPU offload, median of 6 renders (92–104 s)"},{"lane":"video","typical_ms":2102000,"source":"measured on our server 2026-09-23: Wan2.1-T2V-14B in ComfyUI, 832×480, 81 frames (5 s at 16 fps), 50 steps, 2 renders (35:26, 34:38; 41–46 s per step) on GPU0 shared with live services"},{"lane":"video (self-host, LTX-2.5)","typical_ms":62700,"source":"measured on our server 2026-09-23: LTX-2.5 distilled INT8 in ComfyUI, 5 s 1280×704 with audio, 2 renders"},{"lane":"video (self-host, MiniMax-H3)","typical_ms":454000,"source":"measured on our server: MiniMax-H3 t2v 960×544, 124 frames, median of 36 renders, render-studio job records to 2026-09-03 (before the licence review)"}],"benchmark":null,"notes":["MiniMax H3 is self-host only: its licence excludes the US, EU, UK and South Korea, so it never runs in the hosted demo, the gallery or the queue.","LTX-2.5 is self-host only: the LTX-2 Community License bars use in a directly competing hosted service without a commercial licence.","Gallery music from MiniMax-Music3 is labelled \"Powered by MiniMax-Music3\".","Hosted video renders run on MiniMax H3 Max through fal (its MiniMax partnership), billed at fal's list price with no markup: about $0.40 per 5 s 1080P clip. This queue pauses when the provider account or the daily cap runs out, and the console says whether it is open right now; Film and Brand & Ads make video on Decosa's hosted service either way. They need an API key (2 renders per key per day) and stop at a global daily cap. Frames are checked automatically for brand lettering. Running H3 or LTX weights yourself stays a self-host option under their own licences (see above)."]},"buyer_facts":[{"label":"Data retention","value":"Rendered files stay under /studio/media on the server; job status is kept in a file without tokens, and without the prompt once a job has started. Logs carry prompt hashes, never prompts. On fal (hosted video only): fal keeps request data (prompts and file links) for 30 days by default and serves uploaded and generated files from public CDN links until they are deleted [V: fal docs, Data Retention & Storage, read 29 Sep 2026]; Decosa downloads each result at once, and fal's no-storage and short-expiry options are not switched on yet."},{"label":"What leaves the box","value":"Hosted: your prompt reaches Decosa's render service and the text model's prompt check through our gateway; hosted video prompts (text only: no photos, voices or consent recordings) go to fal, a partner GPU provider, under Decosa's account. Self-host: nothing."},{"label":"Inputs","value":"A prompt of up to 1,000 characters (the console takes 600); music takes lyrics, seconds and a seed; images take an aspect (square, landscape, portrait)."},{"label":"Outputs","value":"MP3 songs (MiniMax-Music3 must be credited: 'Powered by MiniMax-Music3'), PNG images, MP4 video; each with a C2PA credential and a signed render receipt."},{"label":"Typical time","value":"Measured: a song in about a minute, an image in a couple of minutes, Wan2.1 video in over half an hour."}],"data_handling":{"page":"/data#studio","self_host":{"level":"confidential","leaves":"nothing","summary":"Runs on your machine; nothing is sent to Decosa or a third party by default."},"hosted":{"level":"operator-processed","demo_only":false,"summary":"TLS to Decosa's server, then decrypted and processed by Decosa's API server, with the open models run by NEAR AI through OpenRouter, with Reka AI as the only fallback under Decosa's account.","gpus":"operator-contracted","third_parties":["fal"],"retention":"Rendered files stay under /studio/media on the server; job status is kept in a file without tokens, and without the prompt once a job has started. Logs carry prompt hashes, never prompts. On fal (hosted video only): fal keeps request data (prompts and file links) for 30 days by default and serves uploaded and generated files from public CDN links until they are deleted [V: fal docs, Data Retention & Storage, read 29 Sep 2026]; Decosa downloads each result at once, and fal's no-storage and short-expiry options are not switched on yet.","used_for_training":false,"encrypted_while_processed":false},"sealed_tier":{"applies":false,"note":"The sealed tier (raw chat only, never use-case pipelines) is paused at launch (/docs/sealed-tier)."},"external_calls":[{"to":"fal (MiniMax H3 Max)","route":"hosted","sends":"content","what":"Hosted video renders send the video prompt (text only) to fal, a partner GPU provider, under Decosa's account, billed at fal's list price with no markup; fal keeps request data 30 days by default. Songs and images render on Decosa's hosted service; self-hosted video renders locally with Wan2.1.","default":"always","off":null}]},"console":{"href":"/studio","input":"gallery","lanes":[{"id":"music","title":"Music","kind":"list"},{"id":"image","title":"Images","kind":"list"},{"id":"video","title":"Video","kind":"list"}],"samples":[{"n":1,"id":"music3-harbor-lights","title":"Music3 harbor lights","deep_link":"/studio?sample=1&autorun=0"},{"n":2,"id":"music3-neon-drive","title":"Music3 neon drive","deep_link":"/studio?sample=2&autorun=0"},{"n":3,"id":"music3-paper-garden","title":"Music3 paper garden","deep_link":"/studio?sample=3&autorun=0"},{"n":4,"id":"music-lofi-evening","title":"Music lofi evening","deep_link":"/studio?sample=4&autorun=0"},{"n":5,"id":"music-orchestral-rise","title":"Music orchestral rise","deep_link":"/studio?sample=5&autorun=0"},{"n":6,"id":"music-indie-morning","title":"Music indie morning","deep_link":"/studio?sample=6&autorun=0"},{"n":7,"id":"music-ambient-tide","title":"Music ambient tide","deep_link":"/studio?sample=7&autorun=0"},{"n":8,"id":"image-lighthouse-dawn","title":"Image lighthouse dawn","deep_link":"/studio?sample=8&autorun=0"},{"n":9,"id":"image-greenhouse-cafe","title":"Image greenhouse cafe","deep_link":"/studio?sample=9&autorun=0"},{"n":10,"id":"image-fox-watercolor","title":"Image fox watercolor","deep_link":"/studio?sample=10&autorun=0"},{"n":11,"id":"image-poster-typography","title":"Image poster typography","deep_link":"/studio?sample=11&autorun=0"},{"n":12,"id":"image-isometric-city","title":"Image isometric city","deep_link":"/studio?sample=12&autorun=0"},{"n":13,"id":"image-macro-dew","title":"Image macro dew","deep_link":"/studio?sample=13&autorun=0"},{"n":14,"id":"video-wan-forest-sunrise","title":"Video wan forest sunrise","deep_link":"/studio?sample=14&autorun=0"},{"n":15,"id":"video-wan-rainy-window","title":"Video wan rainy window","deep_link":"/studio?sample=15&autorun=0"}],"deep_link_params":{"sample":"1-based index into samples, or a sample id","autorun":"1 = start the run once the sample is loaded; 0 (default) = only preselect","reduce-motion":"1 = turn off animations"}},"api":{"base":"https://api.decosa.ai","contract":"/api/contract.json","contract_markdown":"/api/contract.md","reference":"/docs/api","keys":"/account/keys"},"prompts":{"hosted":"/prompts/studio-hosted.md","selfhost":"/prompts/studio-selfhost.md","assemble":"/prompts/studio-assemble.md","mac":"/prompts/studio-mac.md"},"rehearsal":{"bundle":"/samples/studio.zip","bundle_url":"https://decosa.ai/samples/studio.zip","folder":"/samples/studio/","expected":"/samples/studio/expected.json","files":["/samples/studio/expected.json","/samples/studio/inputs/imitation-prompt.json","/samples/studio/inputs/rainy-window-credentialed.mp4","/samples/studio/inputs/rainy-window-one-byte-changed.mp4"],"bytes":964918,"checks":["the gallery lists music, images and video","the gallery has at least one video","every MiniMax-Music3 track carries its licence notice","the render checks as credentialed","its C2PA content hash matches","it leads to the Wan2.1 render receipt","the render receipt's signature verifies","the file with one byte changed checks as tampered","and its content hash no longer matches","the named-artist prompt is refused as imitation (HTTP 422)","no job is created for it"],"licence":"rainy-window-credentialed.mp4: rendered by Decosa with Wan-AI/Wan2.1-T2V-14B (Apache-2.0) from an original prompt; no real people. CC0 for the clip; the copy with one changed byte is the same clip.","about":"One finished studio render (a 5 s Wan2.1 video from the gallery), a copy with one byte changed, and a music prompt that imitates a named artist. The gallery must list music, images and video; the render must check as credentialed with a valid render receipt; the changed copy must check as tampered; and the imitation prompt must be refused (HTTP 422) before any job exists. No new render: renders take the shared GPU for minutes and hosted video is paid.","run":{"containers":"docker compose exec api python scripts/rehearse.py studio","checkout":"python scripts/rehearse.py studio --bundle studio.zip --base-url http://127.0.0.1:8445","mac":".venv/bin/python scripts/rehearse.py studio"},"guidance":"Set up with a coding agent (we recommend Claude Code with Claude Opus 5.5; any capable coding agent works) on mock data only, run the rehearsal until every check passes, then run your own data locally yourself. Never give the agent real data during setup."},"hardware_fit":{"check":"/self-host/hardware?use=studio","data":"/api/hardware.json","tiers":[{"id":"lite","gpu_gb":69.6,"basis":"estimate","unknown":["image-lite"]},{"id":"standard","gpu_gb":50,"basis":"stack","unknown":[]},{"id":"best","gpu_gb":96,"basis":"stack","unknown":[]},{"id":"wanted","gpu_gb":96,"basis":"stack","unknown":[]}],"mac":{"fit":"partial","memory_gb":64}},"links":{"page":"/studio","json":"/use-cases/studio.json","metrics":"/metrics/studio","console":"/studio","console_sample":"/studio?sample=1&autorun=0","stack":"/studio#stack","try_live":"/studio","watch":"/studio","build":"/studio#build","self_host":"/studio#self-host","prompts":{"hosted":"/prompts/studio-hosted.md","selfhost":"/prompts/studio-selfhost.md","assemble":"/prompts/studio-assemble.md","mac":"/prompts/studio-mac.md"}}}