{"id":"shruti","slug":"shruti","name":"Shruti","orgId":"udgam-labs","family":"shruti","releaseDate":null,"availability":"announced","openWeights":true,"license":null,"paramsTotal":null,"paramsSource":null,"contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://udgamlabs.com/blog/shruti","hfRepo":null,"modalities":["audio"],"aliases":["Shruti","shruti","udgamlabs/Shruti","Udgam Shruti"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-10-24","announced":"2026-10-19","blurb":"Udgam Labs' speech-to-text model for Hindi, English and Hinglish in one 29.5 MB file that runs offline on a phone, laptop or in a browser tab. Built on Cactus Compute's Cactus Whistle: Udgam fine-tuned all 55M weights on 2,732 hours of openly licensed Indian speech and added 714 Devanagari vocabulary pieces. Apache-2.0.","api":"none","openWeights":"yes","params":{"text":"55M (29.5 MB file, 4-bit weights / 8-bit activations)","source":"Udgam Labs model page"},"facts":[{"label":"Base model","value":"Cactus-Compute/whistle (Cactus Whistle, Apache-2.0); fine-tuned derivative by Udgam Labs, not trained from scratch"},{"label":"Training data","value":"2,732 hours of openly licensed Indian speech (IndicVoices, Vaani, Shrutilipi, Common Voice 17 and others)"},{"label":"Languages","value":"Hindi in Devanagari, English, and Hinglish with English words kept in Latin script"},{"label":"Runtime","value":"One .cact file for Cactus's open-source needle engine (Mac, Linux, Windows, Android, iOS, WebAssembly); no server needed"},{"label":"Input","value":"16 kHz mono audio; one pass up to 30 s, longer audio in 30 s windows"},{"label":"License","value":"Apache-2.0 (weights), keeping Cactus's notice"}],"claims":["Lab claim: word error rate on Hindi of 20.46 (Monsoon), 19.92 (FLEURS) and 26.28 (Lahaja), against Whisper large-v3's 33.29, 28.88 and 31.19 in Udgam's harness; Udgam reports Whisper large-v3 clearly ahead on English.","Lab claim: 35.21 WER on MUCS 2021 Hinglish, the fewest errors among the open models Udgam measured; the larger Hindi-only Koyal-hi-120M and IndicConformer are more accurate on pure Hindi.","Lab claim: about 1 second per sentence and 80 MB peak memory offline on a Samsung Galaxy F62."],"sources":[{"kind":"lab","title":"Shruti release post (Udgam Labs blog)","url":"https://udgamlabs.com/blog/shruti","date":"2026-10-19"},{"kind":"lab","title":"Shruti model page (Udgam Labs)","url":"https://udgamlabs.com/models/shruti","date":"2026-10-19"},{"kind":"repo","title":"udgamlabs/Shruti model card (Hugging Face)","url":"https://huggingface.co/udgamlabs/Shruti"},{"kind":"repo","title":"udgamlabs/shruti (GitHub)","url":"https://github.com/udgamlabs/shruti"}],"firstBenchmark":null},"organization":{"id":"udgam-labs","name":"Udgam Labs","slug":"udgam-labs","country":null,"url":"https://udgamlabs.com","color":"#11110F","mark":"U"},"price":null,"answer":{"word":"Pending benchmarks.","scope":"Announced Oct 19, 2026; no independent results yet.","lead":"As of Nov 3, 2026, Shruti is pending benchmarks on isitsota: announced Oct 19, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}