{"slug": "soro-a-lightweight-foundation-model-and-chatbot-for-tajik", "title": "Soro: A Lightweight Foundation Model and Chatbot for Tajik", "summary": "Researchers have developed Soro, a family of Tajik-specialized conversational AI models built from Gemma 3 checkpoints and trained on a 1.9-billion-token Tajik corpus. The models outperform same-size baselines on new Tajik benchmarks covering general knowledge and exam domains, while retaining English performance. Soro's quantized versions enable deployment on low-resource devices, supporting an ongoing education pilot in Tajikistan.", "body_md": "arXiv:2605.27379v1 Announce Type: new\nAbstract: We present Soro, a family of Tajik-specialized conversational large language models (LLMs) designed for real-world deployment under tight compute and connectivity constraints in Tajikistan. Starting from open-weight Gemma 3 checkpoints, we perform Tajik-only continual pretraining on a curated 1.9-billion-token corpus spanning filtered web text, PDF documents, and curriculum-aligned educational materials, followed by supervised instruction tuning on 40K Tajik teacher-style examples. To enable rigorous evaluation despite the limited coverage of Tajik in standard benchmarks, we introduce a suite of Tajik benchmarks covering general knowledge, linguistic competence, and school- and university entrance-exam domains, and we open-source them on Hugging Face. Across these Tajik benchmarks, Soro substantially outperforms same-size Gemma 3 baselines while retaining strong English performance on standard datasets. We further show that FP8 and INT4 quantization of Soro preserves most Tajik-language gains while reducing memory requirements for edge deployment, supporting an ongoing education-sector pilot and planned scale-out across schools in Tajikistan.", "url": "https://wpnews.pro/news/soro-a-lightweight-foundation-model-and-chatbot-for-tajik", "canonical_source": "https://arxiv.org/abs/2605.27379", "published_at": "2026-05-28 04:00:00+00:00", "updated_at": "2026-05-28 04:31:07.914121+00:00", "lang": "en", "topics": ["large-language-models", "natural-language-processing", "artificial-intelligence", "machine-learning", "ai-research"], "entities": ["Soro", "Gemma 3", "Hugging Face", "Tajikistan"], "alternates": {"html": "https://wpnews.pro/news/soro-a-lightweight-foundation-model-and-chatbot-for-tajik", "markdown": "https://wpnews.pro/news/soro-a-lightweight-foundation-model-and-chatbot-for-tajik.md", "text": "https://wpnews.pro/news/soro-a-lightweight-foundation-model-and-chatbot-for-tajik.txt", "jsonld": "https://wpnews.pro/news/soro-a-lightweight-foundation-model-and-chatbot-for-tajik.jsonld"}}