Skip to content

Releases: PySpur-Dev/pyspur

v0.0.4

08 Jan 21:37
Compare
Choose a tag to compare
v0.0.4 Pre-release
Pre-release

What's Changed

Full Changelog: v0.0.3...v0.0.4

v0.0.3

07 Jan 00:30
Compare
Choose a tag to compare
v0.0.3 Pre-release
Pre-release

What's Changed

  • Bug fixes
  • refactor: simplify DynamicNode component by removing unused state and… by @srijanpatel in #80

Full Changelog: v0.0.2...v0.0.3

v0.0.2

06 Jan 01:56
b4d68ad
Compare
Choose a tag to compare
v0.0.2 Pre-release
Pre-release

What's Changed

New Contributors

Full Changelog: v0.0.1...v0.0.2

v0.0.1

03 Jan 22:01
Compare
Choose a tag to compare
v0.0.1 Pre-release
Pre-release

PySpur v0.1.0 - First Release

Main features supported are:

  • Create, save and run spurs (workflows/agents) locally on Linux & MacOS

  • Run evals on saved agent workflows

  • Trigger runs via REST API

  • Trace view of runs

  • The following nodes are part of this release:

    • Single LLM Call Node - calls the LLM provider of user's choice in structured output/json mode
    • Best Of N Node - sample the response from LLM n times and pass the best one forward
    • Branch Solve Merge Node - given an input prompt, branch it into parallel steps and then merge the solutions of these steps
    • Router Node - pass the inputs to one of the routes based on conditions defined in the config
    • Coalesce Node - pick the first non-null incoming edge and pass it forward
    • Python Function Node - execute a python function
  • Support for the following LLM providers is part of this release:

    • OpenAI
    • Anthropic
    • Gemini
    • Ollama
      • models such as llama3.2, gemma2, deepseek, mistral, etc can be hosted locally using ollama and be used with pyspur

What's Changed

New Contributors

Full Changelog: https://github.com/PySpur-Dev/pyspur/commits/v0.0.1