Companion evidence for the model checkpoints. This archive is not yet complete. Evidence catch-up was requested on October 15, 2026. Upload status gives dated counts, and catalog.json lists checksum-verified parts and pending evidence groups. A completed local evaluation is not necessarily uploaded yet. Evidence is public. Model uploads and verified local model cleanup resumed on October 15, 2026; see the model checkpoint catalog for model availability. 1. Find the checkpoint in checkpoint-index.json. Entries link original paths, model IDs/revisions, exact matching evaluations, and related training/data units. 2. Find each unit ID in catalog.json. Its parts contain immutable revisions…
SAVRN Model Hub
AI Training Datasets
Each dataset with its card, its structure (every split, row count and column type), its files, its license, and the models that disclose training on it.
11,122Models
3,164Datasets
285Papers
6,792Publishers
11,852Sourced relationships
Updated 2026-10-18 · How the library is built
3,164 datasets, sorted by most downloaded.
Accepted zero-latency teacher rollouts for the supported benchmark tasks. - humanoidbenchbalancesimple: 90 training and 10 validation episodes. The observation.image values are PNG bytes declared as the Hugging Face Image feature, so the Dataset Viewer renders them instead of showing their encoded representation. Canonical LeRobot MP4 files remain under each split's videos/ directory. - mikasainterceptgrabfast: fixed 250/25 StarVLA split. Every Viewer row includes its corresponding observation.images.top and observation.images.wrist frame. The canonical LeRobot data remains under the task path with its RGB observations stored in MP4 files.…
Chaashini — Hindi/Urdu for sugar syrup — is a continuously growing corpus of clean, single-speaker, studio-grade Indian-language speech built for training speech models (text-to-speech, speech recognition, speech language models). Every clip in the corpus has passed a strict multi-stage quality gate; the aim is purity over volume. The corpus grows automatically: new shards are appended every ~2 hours of newly accepted audio. Audio is sourced from publicly available spoken-word recordings (talks, interviews, narration, lectures, podcasts and similar long-form speech). Each recording then passes through: 1. Source-level screening – recordings dominated by music, singing, or non-speech content…
National, first-level subnational and supranational law worldwide (statutes, regulations, gazettes, codes), normalized to the same schema as worldmunicipallaw. This repo holds national, first-level subnational (state / Land / province / région) and supranational law, i.e. places whose placeid is iso:{CC}, iso:{ISO 3166-2} or iso:EU (placekey iso, adminlevel country / adm1 / supranational). Municipal and other local law (municipalities, counties, water boards, …) with the same schema and layout. The data was split out of worldmunicipallaw on 2026-10-11 (copied, verified by sha256 and row counts, then deleted there in a later commit); both repos are now updated by the same pipeline run.…
O RepoJus é um corpus consolidado de decisões judiciais e administrativas brasileiras obtidas de fontes públicas oficiais. O snapshot v2 reúne 4.106.391 registros canônicos, sem duplicatas de normalizedhash, em 4.666 shards Parquet determinísticos (39.290.819.138 bytes). O corpus foi estruturado para pesquisa jurídica, recuperação de informação, RAG, classificação, sumarização, extração de entidades e treinamento ou avaliação
An autonomous audit of small-language-model repos on the Hugging Face Hub. For each in-scope model (independent builders training very small models from scratch, roughly 0.5M–500M parameters), the parameter count stated in the model card is compared against the actual artifact: the safetensors header, config.json, and the training script where present. A mismatch is recorded when the card's number does not match the artifact's real parameter count, or when the card's stated tying / precision / layer structure contradicts the artifact. This is a snapshot. It was generated from the live verification store at 2026-10-03T15:56:31+00:00 by @Compactbot. The live store keeps growing as new models…
Municipal föreskrifter and reglementen (the statutory författningssamling collections) of Swedish kommuner, harvested from the kommuners' own published PDFs and normalized into the worldmunicipal 19-column schema (one row per §/section). documents, 346 rows — see coveragereport.md for the registry, route notes and per-kommun outcomes. The föreskrifter are public legal instruments published by the kommuner under the Kommunallagen; each row carries its source PDF URL. Retrieved (CODE = SCB kommunkod), validation.json sidecars, jurisdictions.jsonl (all 290 kommuner), coveragereport.md, scrapers/ (seprobe.py — site/robots probe; sediscover.py — collection-index discovery; sefetch.py /…
Millions of Burmese sentences synthesized to audio using MyanmarTTS — a from-scratch 31M-parameter flow-matching TTS model (CC0 1.0, pip install myanmartts). Derived from freococo/myanmarspokencorpus, which merges Burmese text from FineWeb2, DCAI, DCAD200, and other web corpora. All files are Parquet, each holding ~5,000 rows: CC0 1.0 — public domain.
clean and sim ready assets - verify visual/collision meshes - modified the collision mesh for better sim - [ ] Robotiq2F85 has vmesh loading problem (which does not happen in PandaRobotiq85 or UR5eRobotiq85) - [ERROR] ValueError: Error: number of faces should be between 1 and 200000 in STL file; perhaps this is an ASCII file? - check the file type by file - if ASCII, check if you have initialized the git LFS - check if the error is resolved.
MLEBench Goal and Flame Chase 6h Campaign Backup Private operational backup of the 2026-09-03 FlowBench-isolated MLEBench campaigns. It preserves the historical Goal and Flame Chase evidence and the fresh 75-task Kimi K3 run, 75 GLM-5.3 Goal cells, and 75 fixed-time GPT-5.6 Sol/Kimi K3 Flame Chase cells. Historical Kimi records do not count toward the fresh continuation result. At this snapshot, the
TTS Arena's DB is SQLlite DB file. The above is just a summary query that should be useful for TTS developers to evaluate faults of their model. Unsafe. Cannot constantly oversee the output of uncontrolled HuggingFace Spaces. While it could be safeguarded by using an ASR model before uploading, something unwanted may still slip through. The one used in dataset viewer. Note that the chosen column may include models that the rejected model beat more times. That is also why votes may sometimes be even less than the amount of distinct chosen models. If you use this data in your publication, please cite us! Copy the BibTeX citation to cite this source
Research snapshot of official legislation collected from Diario da Republica / GUE / Portal do Governo. Not legal advice. Official gazettes / government portals prevail over this corpus. - data/laws.parquet — one row per instrument (id, title, full text, dates, status, identifiers, article count, source URL). - data/articles.parquet — one row per article/section (lawid, article number, title, text, source URL). - scrapers/collectao.py — research collector used to build this snapshot. - scrapers/common.py — shared collector helpers. - scrapers/worldlib.py — shared fetch (live official URL, Wayback fallback). - scrapers/archivefallbacks.py — Wayback / Common Crawl CDX helpers. Official…
Research snapshot of official legislation collected from Federal Negarit Gazette. Not legal advice. Official gazettes / government portals prevail over this corpus. - data/laws.parquet — one row per instrument (id, title, full text, dates, status, identifiers, article count, source URL). - data/articles.parquet — one row per article/section (lawid, article number, title, text, source URL). - scrapers/collectet.py — research collector used to build this snapshot. - scrapers/common.py — shared collector helpers. - scrapers/worldlib.py — shared fetch (live official URL, Wayback fallback). - scrapers/archivefallbacks.py — Wayback / Common Crawl CDX helpers. Official government / gazette texts…
Opaque authenticated ciphertext managed by HF Drive. No plaintext files or keys here.
Model Collections
Hand-picked starting points, each with the reason it exists.
Collection · 4 entries
Embedding models for retrieval
Sentence and document embedding models used to build retrieval systems. Dimension and sequence length matter more than size here, and both come from the publisher.
Collection · 4 entries
Models that fit on one accelerator
Models whose publisher-reported parameter count puts them within reach of a single accelerator at common precisions. Memory needed depends on precision and serving configuration, so treat the parameter count as the starting point, not the answer.
Collection · 6 entries
Open-weight text models worth knowing
Widely used open-weight language models, chosen because each one is a distinct family rather than a variant of the one above it. Selection, not a ranking.
Collection · 3 entries
Speech and audio models
Recognition and synthesis models, grouped so the two directions are easy to compare.
Related SAVRN Research
The hub sits beside SAVRN's market data and infrastructure research: what models cost to run, and what it takes to run them.
SAVRN Index
What open models cost to run
The same open-weight model priced by every host that serves it, per million tokens.
Research Hub
Data center trackers and maps
Moratoriums, permits, power, water and capital behind the facilities that run these models.
Method
How the Model Hub is built
Sources, evidence labels, refresh behaviour, and the limits of every comparison here.