Zipstack / Zipstack/unstract

feat: Parallel exection of prompts in Prompt Studio

Open
#878 1 comment 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

feature
Dominant language
Python
Stars
7.2k
Forks
718
Avg merge
3d 1h
Merged PRs (30d)
23

Description

Is your feature request related to a problem? Please describe.
Currently prompts defined in a Prompt Studio are executed sequentially by default. This drastically increases the latency as the number of defined prompts increases.

Describe the solution you'd like

  • The defined prompts need to be executed in parallel instead for obvious latency improvements.
  • Might be better to add a toggle button to choose whether to run all prompts in parallel or not.

Describe alternatives you've considered

  • SinglePassExtraction might not be feasible if expected output and/or passed context is big for each prompt.
  • SummarizedExtraction might not be feasible if each clause needs to be analyzed (e.g. insurance contracts documents).

Additional context
Credits to Michel Medawar.
See original public Slack thread.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by locating the Prompt Studio entry point that executes defined prompts and trace how prompts are currently run sequentially. Determine whether parallel execution alone or a user-facing toggle is required, then verify that all defined prompts complete correctly with the intended latency improvement.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
ai
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.