Open Models Almanac
Methodsinference & servingagentic scaffolding

single mention · inference & serving

StreamableParser

A StreamableParser parses and decodes harmony messages incrementally as the model generates new tokens.1 It processes each streamed token and exposes the current role, channel, recipient, and content, enabling streamed output and correct unicode handling during decoding.2

How this definition was written

Summarised by glm-5.3-flash from 1 source; every sentence is checked against the quote it cites.

  1. a StreamableParser for parsing and decoding as the model is generating new tokens — OpenAI Harmony Response Format
  2. This can be helpful for example to stream output and handle unicode characters during decoding — OpenAI Harmony Response Format

Filed under inference & serving › agentic scaffolding