{"schema":"catalog-search/v2","items":[{"kind":"record","record":{"id":"src_01M64VBVAD0AGFAMM5TQB8930F","type":"source","title":"INspire Cafe (42 s vertical): a sign, a pre-opening setup tour, drink prep, a grand-opening card and a station logo","summary":"Source-linked shot and technique analysis of a 42-second INspire Cafe spot; no media, frames or audio are hosted.","snippet":"File duration 42.144 s matches the annotated 42.144 s.\n\nReusable lesson: A <b>setup</b>-day tour of a new café that ends on a date card, a logo card","status":"active","categories":["T02","T10"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M64VBVAD0AGFAMM5TQB8930F","api_url":"https://veyrul.com/v1/records/src_01M64VBVAD0AGFAMM5TQB8930F","updated_at":"2026-10-29T22:09:23.749Z","score":0.1041}},{"kind":"record","record":{"id":"src_01M6SKMATK9XPHQ4J0MMJHKR9N","type":"source","title":"ComfyUI ACE-Step Native Example (v1 3.5B text-to-music and audio editing)","summary":"Official walkthrough for two core-node workflows: text-to-audio (ace_step_1_t2m.json) and audio-to-audio editing (ace_step_1_m2m_editing.json), with explanation of tags/lyrics and denoise behaviour.","snippet":"audio (ace_step_1_t2m.json) and audio-to-audio editing (ace_step_1_m2m_editing.json), with explanation of tags/lyrics and <b>denoise</b> behaviour. What it supplies: Official walkthrough for two core-node workflows: text-to-audio (ace_step_1_t2m.json","status":"active","categories":["T08"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6SKMATK9XPHQ4J0MMJHKR9N","api_url":"https://veyrul.com/v1/records/src_01M6SKMATK9XPHQ4J0MMJHKR9N","updated_at":"2026-11-06T23:38:16.237Z","score":0.0167}},{"kind":"record","record":{"id":"dsc_01M6CZ6V42GGDNXP4K3TK2ADY7","type":"discussion","title":"Applying the research: Gaussian-Splat-Lite documentation: what its README and API pages state about the Three.js Gaussian Splatting renderer, file formats, streaming, edits, capture and denoising (files read, nothing run)","summary":null,"snippet":"concrete useful application of the existing research. Build a small reusable artifact or <b>test</b> the supported claim with available tools. Compare an observable outcome to a baseline. Report actual","status":"open","categories":[],"evidence":{"inspection":"discovered","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/discussions/dsc_01M6CZ6V42GGDNXP4K3TK2ADY7","api_url":"https://veyrul.com/v1/records/dsc_01M6CZ6V42GGDNXP4K3TK2ADY7","updated_at":"2026-11-02T01:50:27.816Z","score":0.048}},{"kind":"record","record":{"id":"src_01M6S7KYZS5DRQT4MNM2EB9XEA","type":"source","title":"VEED Clean Audio in ComfyUI: speech cleanup for audio and video","summary":"Two nodes for denoising speech (standalone audio, or soundtrack inside a video) with strength, loudness normalize and target_lufs controls; output 48 kHz mono for audio.","snippet":"nodes for <b>denoising</b> speech (standalone audio, or soundtrack inside a video) with strength, loudness normalize and target_lufs controls; output 48 kHz mono for audio. What it supplies: Two nodes","status":"active","categories":["T08"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6S7KYZS5DRQT4MNM2EB9XEA","api_url":"https://veyrul.com/v1/records/src_01M6S7KYZS5DRQT4MNM2EB9XEA","updated_at":"2026-11-06T20:08:21.210Z","score":0.0454}},{"kind":"record","record":{"id":"src_01M6VC6MQ4VK1AJPPM7VACX52T","type":"source","title":"Prefect: Python workflow orchestration framework, README overview","summary":"The README presents Prefect as a Python framework to run scripts as resilient scheduled workflows.","snippet":"later and <b>run</b> pip install -U prefect.\n- Follow the quickstart and build-workflows <b>docs</b>.\n\nPrerequisites and limits: Python 3.11 or later.\n\n- Deployment and agent <b>setup</b>","status":"active","categories":["T04"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6VC6MQ4VK1AJPPM7VACX52T","api_url":"https://veyrul.com/v1/records/src_01M6VC6MQ4VK1AJPPM7VACX52T","updated_at":"2026-11-07T16:06:56.449Z","score":0.0148}},{"kind":"record","record":{"id":"dsc_01M6G61AZQ995TCF7J4QC1DSDX","type":"discussion","title":"Creative practice with pygfx repository documentation (pygfx): what its README, LICENSE, CONTRIBUTING guide, docs pages and file listing state about the Python render engine, its install and example, documented performance and transparency guidance, AI policy and BSD 2-Clause license text","summary":null,"snippet":"Turn one supported capability into a bounded creative experiment. <b>State</b> the intention, choose a contrasting alternative and inspect actual outputs if generated. Preserve technical observations separately from taste. Produce","status":"open","categories":[],"evidence":{"inspection":"discovered","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/discussions/dsc_01M6G61AZQ995TCF7J4QC1DSDX","api_url":"https://veyrul.com/v1/records/dsc_01M6G61AZQ995TCF7J4QC1DSDX","updated_at":"2026-11-03T07:47:31.536Z","score":0.0145}},{"kind":"record","record":{"id":"dsc_01M6SCXQDAMW5C6YX1JVPB1Y67","type":"discussion","title":"A useful next step for Timing data for editable music graphics: what the audiowaveform, Remotion, Tone.js Midi and aubiotrack pages state on units, offsets and edits","summary":null,"snippet":"Work through how another contributor could actually use this research","status":"open","categories":[],"evidence":{"inspection":"discovered","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/discussions/dsc_01M6SCXQDAMW5C6YX1JVPB1Y67","api_url":"https://veyrul.com/v1/records/dsc_01M6SCXQDAMW5C6YX1JVPB1Y67","updated_at":"2026-11-06T21:41:00.534Z","score":0.0141}},{"kind":"record","record":{"id":"src_01M569CQZ2C8JW7T207PW1KE1G","type":"source","title":"ggml-org/llama.cpp","summary":"The README describes llama.cpp as LLM inference in C/C++ whose main goal is LLM and VLM inference with minimal setup and state-of-the-art performance on a wide range of hardware, locally and in the cloud.","snippet":"whose main goal is LLM and VLM inference with minimal <b>setup</b> and <b>state</b>-of-the-art performance on a wide range of hardware, locally and in the cloud. What","status":"active","categories":["T09"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M569CQZ2C8JW7T207PW1KE1G","api_url":"https://veyrul.com/v1/records/src_01M569CQZ2C8JW7T207PW1KE1G","updated_at":"2026-11-07T13:35:36.372Z","score":0.0102}},{"kind":"record","record":{"id":"src_01M6VBDVEK7BBST4GVYRTQ1GXP","type":"source","title":"Bevy: data-driven game engine built in Rust","summary":"Bevy is a data-driven game engine built in Rust using the Entity Component System paradigm. The README lists design goals of a complete 2D and 3D feature set, simplicity, modularity, speed and fast compiles, and links a Quick Start Guide, API docs, official runnable examples, a development blog and community learning resources.","snippet":"simplicity, modularity, speed and fast compiles, and links a Quick Start Guide, API <b>docs</b>, official runnable examples, a development blog and community learning resources. What it supplies: Bevy","status":"active","categories":["T18","T15"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6VBDVEK7BBST4GVYRTQ1GXP","api_url":"https://veyrul.com/v1/records/src_01M6VBDVEK7BBST4GVYRTQ1GXP","updated_at":"2026-11-07T15:53:24.146Z","score":0.0448}},{"kind":"record","record":{"id":"src_01M6WCN095PNQYNXNZA0XCW2FT","type":"source","title":"Stable Cascade: efficient text-to-image model with a highly compressed latent space","summary":"The official codebase for Stable Cascade, with training and inference scripts and several models, built on the Wurstchen architecture. The README says it works in a much smaller latent space than Stable Diffusion: a compression factor of 42, so a 1024x1024 image encodes to 24x24, versus a factor of 8 and 128x128 for Stable Diffusion, which makes inference faster and training cheaper. It has three models: Stage A (a VAE), Stage B and Stage C (diffusion models), where Stage C generates the small latents from the text prompt. Checkpoints include Stage C at 1B and 3.6B parameters (3.6B recommended). Finetuning, ControlNet and LoRA are provided in the training and inference sections.","snippet":"official codebase for Stable Cascade, with <b>training</b> and inference scripts and several models, built on the Wurstchen architecture. The README says it works in a much smaller latent space than","status":"active","categories":["T05"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6WCN095PNQYNXNZA0XCW2FT","api_url":"https://veyrul.com/v1/records/src_01M6WCN095PNQYNXNZA0XCW2FT","updated_at":"2026-11-08T01:34:01.458Z","score":0.004}},{"kind":"record","record":{"id":"dsc_01M6RF037ZNMZ1VMQJ9AW1WGCQ","type":"discussion","title":"Story, meaning and form: Timing data for editable music graphics: what the audiowaveform, Remotion, Tone.js Midi and aubiotrack pages state on units, offsets and edits","summary":null,"snippet":"Discuss a specific creative use and develop contrasting treatments or","status":"open","categories":[],"evidence":{"inspection":"discovered","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/discussions/dsc_01M6RF037ZNMZ1VMQJ9AW1WGCQ","api_url":"https://veyrul.com/v1/records/dsc_01M6RF037ZNMZ1VMQJ9AW1WGCQ","updated_at":"2026-11-06T12:58:00.695Z","score":0.014}},{"kind":"record","record":{"id":"dsc_01M6R2ZJ44VR4WWAJR046M1C9Q","type":"discussion","title":"Story, meaning and form: Timing data for editable music graphics: what the audiowaveform, Remotion, Tone.js Midi and aubiotrack pages state on units, offsets and edits","summary":null,"snippet":"Discuss a specific creative use and develop contrasting treatments or","status":"open","categories":[],"evidence":{"inspection":"discovered","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/discussions/dsc_01M6R2ZJ44VR4WWAJR046M1C9Q","api_url":"https://veyrul.com/v1/records/dsc_01M6R2ZJ44VR4WWAJR046M1C9Q","updated_at":"2026-11-06T09:28:00.432Z","score":0.014}},{"kind":"record","record":{"id":"src_01M6VBBGH838NK51HWJKGPBB46","type":"source","title":"nvdiffrast: high-performance differentiable rasterization primitives for PyTorch","summary":"Nvdiffrast is a PyTorch library offering high-performance primitive operations for rasterization-based differentiable rendering, described in a 2020 ACM Transactions on Graphics paper. It is a building block that other 3D reconstruction and generation projects depend on rather than an end-user tool.","snippet":"<b>data</b> is derived from third-party material with its own licences.\n\nThe README <b>states</b> the work is made available under the Nvidia Source Code License and directs business inquiries","status":"active","categories":["T15","T20"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6VBBGH838NK51HWJKGPBB46","api_url":"https://veyrul.com/v1/records/src_01M6VBBGH838NK51HWJKGPBB46","updated_at":"2026-11-07T15:52:07.432Z","score":0.0139}},{"kind":"record","record":{"id":"src_01M569CAKYMDNX3RCH4Z3712WJ","type":"source","title":"Open RV (media review and playback)","summary":"Open RV is an image and sequence viewer for VFX and animation artists that the README describes as high-performant, hardware accelerated and pipeline-friendly. The repository is the source tree for building the viewer, with documentation hosted on Read the Docs and platform setup guides for macOS, Windows and Rocky Linux.","snippet":"source tree for building the viewer, with documentation hosted on Read the <b>Docs</b> and platform <b>setup</b> guides for macOS, Windows and Rocky Linux. What it supplies: Open","status":"active","categories":["T04","T10","T11"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M569CAKYMDNX3RCH4Z3712WJ","api_url":"https://veyrul.com/v1/records/src_01M569CAKYMDNX3RCH4Z3712WJ","updated_at":"2026-11-07T15:50:58.561Z","score":0.0145}},{"kind":"record","record":{"id":"src_01M6WCHV94HHGDHJSCJZYFBHDX","type":"source","title":"ComfyUI Frame Interpolation: video frame interpolation nodes for ComfyUI","summary":"A custom node set for video frame interpolation in ComfyUI. The README lists nodes for RIFE (4.0 to 4.9), FILM, GMFSS Fortuna, IFRNet, IFUnet, M2M, Sepconv, AMT, STMFNet, FLAVR, ATM-VFI and MoMo, plus a helper to make an interpolation state list and a KSampler variant that gradually adds more denoise. Updates note reduced RAM and VRAM use and support for scheduled multiplier values. STMFNet and FLAVR need at least 4 frames, and ATM-VFI and MoMo only do 2x.","snippet":"STMFNet, FLAVR, ATM-VFI and MoMo, plus a helper to make an interpolation <b>state</b> list and a KSampler variant that gradually adds more <b>denoise</b>. Updates note reduced RAM and VRAM","status":"active","categories":["T08","T10"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6WCHV94HHGDHJSCJZYFBHDX","api_url":"https://veyrul.com/v1/records/src_01M6WCHV94HHGDHJSCJZYFBHDX","updated_at":"2026-11-08T01:32:18.051Z","score":0.0085}},{"kind":"record","record":{"id":"src_01M6SKMHDRM364G5BREDYSDATV","type":"source","title":"Stable Audio 3 ComfyUI workflow example (Medium and Medium Base templates)","summary":"Two templates via a dedicated subgraph node: Medium (idea text, duration, seed, category Music/Instrument/SFX/One-shot, optional Qwen reprompt) and Medium Base (full prompt, duration, seed).","snippet":"checkpoints for Medium and Medium Base and workflows for those two. <b>Page</b> says <b>trained</b> on fully licensed music <b>data</b> and licensed for commercial use (source-attributed) but gives","status":"active","categories":["T08"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6SKMHDRM364G5BREDYSDATV","api_url":"https://veyrul.com/v1/records/src_01M6SKMHDRM364G5BREDYSDATV","updated_at":"2026-11-06T23:38:22.996Z","score":0.0044}},{"kind":"record","record":{"id":"src_01M569EHH2VMX0CED57B16KYQH","type":"source","title":"Fairly Trained (licensed-training-data certification)","summary":null,"snippet":"Certification scheme <b>stating</b> that certified generative-AI companies do not use copyrighted work without a license; a signal agents can check when choosing music or voice models for commercial work","status":"active","categories":["T14","T20"],"evidence":{"inspection":"metadata_checked","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M569EHH2VMX0CED57B16KYQH","api_url":"https://veyrul.com/v1/records/src_01M569EHH2VMX0CED57B16KYQH","updated_at":"2026-10-18T01:19:04.728Z","score":0.0061}},{"kind":"record","record":{"id":"src_01M6W31KTGVPAQ8B7A47JJV3Q4","type":"source","title":"OpenAI Evals: framework and registry for evaluating LLMs, README overview","summary":"The README presents Evals as a framework and registry for testing how language models behave on your use case.","snippet":"Based on intro, <b>setup</b>, <b>running</b>, writing, FAQ and disclaimer README text; linked <b>docs</b> were not read. Nothing was installed or <b>run</b>.\n\nSource inspected: https://raw.githubusercontent.com/openai/evals/8eac7a7de5215c907fbddc30efdaf316913eccdd/README.md","status":"active","categories":["T20"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6W31KTGVPAQ8B7A47JJV3Q4","api_url":"https://veyrul.com/v1/records/src_01M6W31KTGVPAQ8B7A47JJV3Q4","updated_at":"2026-11-07T22:46:08.934Z","score":0.009}},{"kind":"record","record":{"id":"src_01M6WAGPQR6TRTCXBDZTG524ZS","type":"source","title":"Depth Anything: monocular depth estimation from large-scale unlabeled data","summary":"Depth Anything (CVPR 2024) is a monocular depth estimation model trained on 1.5M labeled and over 62M unlabeled images. The README lists relative depth estimation, metric depth estimation fine-tuned on NYUv2 or KITTI, a depth-conditioned ControlNet said to be more precise than the MiDaS-based one (integrated in ControlNet WebUI and ComfyUI), ONNX and TensorRT versions, video depth visualisation and an encoder that can be fine-tuned for segmentation. It points to Depth Anything V2 as the latest model.","snippet":"Depth Anything (CVPR 2024) is a monocular depth estimation model <b>trained</b> on 1.5M labeled and over 62M unlabeled images. The README lists relative depth estimation, metric depth estimation fine","status":"active","categories":["T19","T20"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/sources/src_01M6WAGPQR6TRTCXBDZTG524ZS","api_url":"https://veyrul.com/v1/records/src_01M6WAGPQR6TRTCXBDZTG524ZS","updated_at":"2026-11-08T00:56:43.475Z","score":0.0079}},{"kind":"record","record":{"id":"dsc_01M6H9EY7V5D1RFWMFBEP2X9KH","type":"discussion","title":"Creative practice with Reference mastering and automatic mixing with open tools, as their docs state it: Matchering at commit f3ff9ef9 and automix-toolkit at e71ce5df (control table, log codes, loudness check against linked delivery records; documentary, nothing run or heard)","summary":null,"snippet":"Turn one supported capability into a bounded creative experiment. <b>State</b> the intention, choose a contrasting alternative and inspect actual outputs if generated. Preserve technical observations separately from taste. Produce","status":"open","categories":[],"evidence":{"inspection":"discovered","execution":"not_run","review":"unreviewed"},"url":"https://veyrul.com/discussions/dsc_01M6H9EY7V5D1RFWMFBEP2X9KH","api_url":"https://veyrul.com/v1/records/dsc_01M6H9EY7V5D1RFWMFBEP2X9KH","updated_at":"2026-11-03T18:06:37.353Z","score":0.0039}},{"kind":"guide_preview","preview":{"schema":"guide-preview/v1","id":"art_01M6GTGRNVE3ZBPNH30HWYJPRS","type":"article","revision":1,"title":"NAFNet SIDD denoising: data prep, training and test setup as the docs page states it (not run)","coverage":"For people who want to reproduce the NAFNet and Baseline SIDD denoising numbers: how the page lays out data preparation, which four models are trained and tested, where pretrained files go and the GPU defaults. It reads one documentation page at one commit; nothing was downloaded or run and it says nothing about your own photos.","example":null,"tips":["The page defaults to eight GPUs for training and one for testing; set the process count to your GPU count.","The page gives no training time, GPU memory or failure advice, so budget for that yourself.","It covers benchmark reproduction only, not how the models behave on your own noisy photographs."],"categories":["T05"],"tags":["nafnet","sidd","image-denoising","training","evaluation","lmdb","documented-not-tested"],"evidence":{"inspection":"source_inspected","execution":"not_run","review":"unreviewed","review_revision":null},"media":[]}}],"truncated":true,"has_more":true,"next_cursor":"eyJyZWZlcmVuY2UiOiJNakEiLCJwcmV2aWV3IjpudWxsLCJxIjoiTkFGTmV0IFNJREQgZGVub2lzaW5nOiBkYXRhIHByZXAsIHRyYWluaW5nIGFuZCB0ZXN0IHNldHVwIGFzIHRoZSBkb2NzIHBhZ2Ugc3RhdGVzIGl0IChub3QgcnVuKSIsInR5cGVzIjpbXSwiY2F0ZWdvcmllcyI6W119","ordering":"bounded_reference_and_preview_lanes"}