LLM Reference

Chat Bison

About

Chat Bison, a large language model from Google, is adept at multi-turn conversations, excelling in language understanding and generation for interactive applications. Although details about its architecture are limited, it is based on Google's PaLM 2 framework and trained on extensive text and code datasets. This model is proficient in various language tasks, including code and text generation, editing, and problem-solving, and is particularly well-suited for multilingual content and handling synonymous inquiries in FAQ systems. Accessible via Google's Vertex AI platform, users should note the scheduled decommissioning of the PaLM API, including Chat Bison, and consider transitioning to the Gemini API.

Capabilities

MultimodalFunction CallingTool UseJSON Mode

Providers(1)

ProviderInput (per 1M)Output (per 1M)Type
GCP Vertex AI
Serverless

Specifications

FamilyPalm 2
Released2023-12-06
ArchitectureDecoder Only
Specializationgeneral