{"paper":{"title":"GigaChat Family: Efficient Russian Language Modeling Through Mixture of Experts Architecture","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ainur Israfilova, Aleksandr Proshunin, Alena Fenogenova, Arkadiy Shatenov, Artem Chervyakov, Daniil Smirnov, Daria Khomich, Darya Latortseva, Dmitry Kozlov, Dzmitry Menshykh, Eldar Damirov, Emil Shakirov, Evgenii Kosarev, Fedor Minkin, GigaChat team: Mamedov Valentin, Grafov Sergei, Gregory Leleytner, Ilya Shchuckin, Ivan Baskov, Karlov Vladimir, Kolodin Egor, Lukyanenko Ivan, Mikhail Kolesov, Nikita Savushkin, Oleg Kutuzov, Polina Kudriavtseva, Ruslan Gaitukiev, Sergei Averkiev, Sergei Porkhun, Sofiia Soldatova, Stanislav Pyatkin, Valeriy Berezovskiy, Yury Fedorov","submitted_at":"2025-06-11T06:46:49Z","abstract_excerpt":"Generative large language models (LLMs) have become crucial for modern NLP research and applications across various languages. However, the development of foundational models specifically tailored to the Russian language has been limited, primarily due to the significant computational resources required. This paper introduces the GigaChat family of Russian LLMs, available in various sizes, including base models and instruction-tuned versions. We provide a detailed report on the model architecture, pre-training process, and experiments to guide design choices. In addition, we evaluate their per"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09440","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2506.09440/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}