{"id":"lilavati","slug":"lilavati","name":"Lilavati","orgId":"udgam-labs","family":"lilavati","releaseDate":null,"availability":"announced","openWeights":true,"license":null,"paramsTotal":null,"paramsSource":null,"contextWindow":null,"knowledgeCutoff":null,"modelCardUrl":"https://udgamlabs.com/blog/lilavati","hfRepo":null,"modalities":["text"],"aliases":["Lilavati","lilavati","udgamlabs/Lilavati","Udgam Lilavati"],"aliasNote":null,"origin":"tracked","tracked":{"trackedSince":"2026-10-24","announced":"2026-10-16","blurb":"Udgam Labs' 95M-parameter English model that solves grade-school maths word problems by writing a short Python program, which is run in a sandbox to give the answer. Trained from scratch and released under Apache-2.0 with Lilavati-Checker, a 95M companion model that rates programs so many tries can be combined.","api":"none","openWeights":"yes","params":{"text":"95M (94.7M; 12 layers, width 640)","source":"Udgam Labs model page"},"facts":[{"label":"Input → output","value":"English word problem → a Python function whose return value is the answer"},{"label":"Training","value":"Trained from scratch on public synthetic maths datasets with Python solutions (TinyGSM and a TinyGSM-style set), general English text, the GSM8K training split and LLM-generated problems kept only if their solutions executed"},{"label":"Context","value":"4,096 tokens"},{"label":"Companion model","value":"Lilavati-Checker (95M, same architecture and tokenizer) estimates how likely each program is right; the lilavati package runs 8 tries by default with a checker-weighted vote"},{"label":"Release","value":"udgamlabs/Lilavati and udgamlabs/Lilavati-Checker on Hugging Face; code, sandbox and evaluation on GitHub (udgamlabs/lilavati)"},{"label":"License","value":"Apache-2.0 (weights and code)"}],"claims":["Lab claim: GSM8K test 54.7% with one greedy try (program executed); 72.3% with the checker-weighted vote over 48 tries.","Lab claim: SVAMP 69.9%, ASDiv (numeric) 74.0%, GSM-Hard 43.7%, GSM-Plus testmini 39.2% with one greedy try.","Lab claim: about 2.7× the GSM8K score of the strongest other open model up to 150M parameters that Udgam tested; Udgam notes TinyGSM's unreleased 125M models reported higher numbers under their own protocol."],"sources":[{"kind":"lab","title":"Lilavati, a 95M model that solves maths word problems by writing Python (Udgam Labs blog)","url":"https://udgamlabs.com/blog/lilavati","date":"2026-10-16"},{"kind":"lab","title":"Lilavati model page (Udgam Labs)","url":"https://udgamlabs.com/models/lilavati","date":"2026-10-16"},{"kind":"repo","title":"udgamlabs/Lilavati model card (Hugging Face)","url":"https://huggingface.co/udgamlabs/Lilavati"},{"kind":"repo","title":"udgamlabs/Lilavati-Checker model card (Hugging Face)","url":"https://huggingface.co/udgamlabs/Lilavati-Checker"},{"kind":"repo","title":"udgamlabs/lilavati (GitHub)","url":"https://github.com/udgamlabs/lilavati"}],"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 16, 2026; no independent results yet.","lead":"As of Nov 3, 2026, Lilavati is pending benchmarks on isitsota: announced Oct 16, 2026, with no independent results on the boards we track yet.","confidence":"low"},"status":"PENDING_BENCHMARKS","verdicts":[]}