{"slug": "umartransit-1b-first-open-source-transit-domain-llm-fine-tuned-qwen2-5-1-5b", "title": "UmarTransit-1B: First Open-Source Transit Domain LLM (Fine-tuned Qwen2.5-1.5B)", "summary": "UmarTransit-1B, the first open-source large language model fine-tuned for public transit systems and GTFS data, has been released. Built by fine-tuning Qwen2.5-1.5B-Instruct using QLoRA, the model specializes in GTFS specification knowledge, route analysis, and stop information across 10 countries. The project is fully open source under Apache 2.0, with a detailed build guide aimed at engineers new to AI/ML.", "body_md": "Hi everyone!\n\nI’m sharing **UmarTransit-1B** — a domain-specific language model fine-tuned for public transit systems and GTFS (General Transit Feed Specification) data.\n\nA QLoRA fine-tuned version of Qwen2.5-1.5B-Instruct that specializes in GTFS specification knowledge, transit route/schedule analysis, stop/station information across 10 countries, and transfer connections.\n\n**Available formats:**\n\nI’m a Senior Technical Lead (Java/Spring Boot, 14+ years) experienced with LLM APIs (OpenAI, Claude) and GTFS/transit systems. This is my first end-to-end model training project — from data collection to fine-tuning to deployment. Built the entire pipeline:\n\nTraining: QLoRA rank=16, alpha=32, 3 epochs, ~30 min on Colab T4, Unsloth + TRL SFTTrainer\n\nQ: What does route_type 3 mean in GTFS? A: In the GTFS specification, route_type 3 indicates Bus service. This is the most common route type and covers short- and long-distance bus routes.\n\nQ: How many routes does the Chicago Transit Authority operate? A: The Chicago Transit Authority (CTA) operates 133 routes. 8 are Subway/Metro routes and 125 are Bus routes.\n\nQ: Can GTFS times exceed 24:00:00? A: Yes, GTFS allows times greater than 24:00:00 in stop_times.txt. This is used for trips that extend past midnight relative to the service day.\n\nThe entire project is open source (Apache 2.0) with a detailed build guide for engineers new to AI/ML. Feedback and contributions welcome!", "url": "https://wpnews.pro/news/umartransit-1b-first-open-source-transit-domain-llm-fine-tuned-qwen2-5-1-5b", "canonical_source": "https://discuss.huggingface.co/t/umartransit-1b-first-open-source-transit-domain-llm-fine-tuned-qwen2-5-1-5b/177611#post_1", "published_at": "2026-07-09 07:06:46+00:00", "updated_at": "2026-07-09 07:21:49.368100+00:00", "lang": "en", "topics": ["large-language-models", "artificial-intelligence", "ai-products", "ai-research"], "entities": ["UmarTransit-1B", "Qwen2.5-1.5B-Instruct", "QLoRA", "GTFS", "Unsloth", "TRL SFTTrainer", "Chicago Transit Authority"], "alternates": {"html": "https://wpnews.pro/news/umartransit-1b-first-open-source-transit-domain-llm-fine-tuned-qwen2-5-1-5b", "markdown": "https://wpnews.pro/news/umartransit-1b-first-open-source-transit-domain-llm-fine-tuned-qwen2-5-1-5b.md", "text": "https://wpnews.pro/news/umartransit-1b-first-open-source-transit-domain-llm-fine-tuned-qwen2-5-1-5b.txt", "jsonld": "https://wpnews.pro/news/umartransit-1b-first-open-source-transit-domain-llm-fine-tuned-qwen2-5-1-5b.jsonld"}}