Integrating GigaChat LLM into a virtual assistant: technical implementation
summary
Describes integrating the GigaChat large language model into Sber’s virtual assistant to enhance its capabilities; covers why an LLM is needed in the assistant and how external skills (third-party integrations) are used. Presents technical implementation details of embedding GigaChat into the assistant’s architecture and practical considerations from the integration effort.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 1:28What is GigaChat LLM
- 3:14Integration challenges
- 5:02LM approach for skills
- 6:51User request handling
- 8:17Skill classifier basics
- 11:52Model roles in dialogue
- 14:40Prompt structure
- 18:29Model training
- 21:06Implementation results
- 22:57Q&A
related talks
43mTraining GigaChat MAX
51mChatGPT as a developer's personal servant
31mA Hybrid LLM based Enterprise Chatbot Solution Weaver Peer by Intellya
27mWHAT DOES MLOps LOOK LIKE FOR LARGE LANGUAGE MODELS (LLMs) IN PRODUCTION?
10mBuilding Conversational AI Agents at Scale
56mLarge Language Models for Python Specialists
47mOperationalization of AI/ML models and use cases for ChatGPT Q&A
51mA ChatGPT on your own data: how does it work?