v0.0.19

What's Changed

• Initial Gemini implementation by @aflament in #149
• Feature add initial support for images by @aflament in #150
• 0.0.19 Release by @aflament in #151

Full Changelog: v0.0.18...v0.0.19

v0.0.18

What's Changed

• Feature code clean up 20240529 by @aflament in #141
• Fix match labels by @aflament in #142
• Feature refactor llm configuration by @aflament in #144
• Feature support new openai models 20240607 by @aflament in #145
• Feature LLM prompt config by @Winston-503 in #146
• Add LLMFunction by @aflament in #147
• added pathway and an easier code for cloning by @Budalebah in #143
• Release 0.0.18 by @aflament in #148

New Contributors

• @Budalebah made their first contribution in #143

Full Changelog: v0.0.17...v0.0.18

v0.0.17

What's Changed

• Add COUNCIL_PYTHON_EXECUTABLE optional env var by @gkoch78 in #131
• fix f string in LLMException by @gkoch78 in #132
• Feature add value validator by @aflament in #135
• Code clean-up by @aflament in #136
• Fix issue to support Anthropic in get_llm_from_config function by @ethancjackson in #134
• Support custom OpenAI api hosts by @pixelkaiser in #133
• Fix integration tests 20240314 by @aflament in #137
• Feature add support for new anthropic models by @aflament in #138
• Feature release 0.17 by @aflament in #139

New Contributors

• @ethancjackson made their first contribution in #134
• @pixelkaiser made their first contribution in #133

Full Changelog: v0.0.16...v0.0.17

v0.0.16

What's Changed

• CU-8686yqa3h_Fix-execute_log_to_json-when-a-log-entry-contains-an-error by @gkoch78 in #123
• Feature add openai new models 20240125 by @aflament in #125
• Add parameters to google skills init by @zikunukiz in #122
• Suppress roadmap section until further notice by @Winston-503 in #124
• Feature improve yaml parsing in llmanswer by @gkoch78 in #127
• Fix timeout by @aflament in #128
• Add ability to configure LLM with Yaml files by @aflament in #129
• Version 0.0.16 by @aflament in #130

New Contributors

• @zikunukiz made their first contribution in #122
• @Winston-503 made their first contribution in #124

Full Changelog: v0.0.15...v0.0.16

v0.0.15

What's Changed

• Feature add anthropic llm by @gkoch78 in #109
• Fix imports from Anthropic by @gkoch78 in #110
• Feature add llm call timeout exception by @aflament in #111
• Feature add wikipedia search skill by @gkoch78 in #112
• Feature model parameters refactoring by @aflament in #113
• Requirements Updates by @aflament in #114
• Improve logging by @aflament in #115
• Feature update notebooks by @gkoch78 in #116
• Feature release 0.0.15 by @aflament in #117

Full Changelog: v0.0.14...v0.0.15

v0.0.14

What's Changed

• Add relevant execution graph node in the execution log entry by @gkoch78 in #101
• Add commonly used types in council main module by @gkoch78 in #102
• Feature improve controller extensibility by @aflament in #103
• Misc code clean-up by @aflament in #104
• Feature add python code skills by @gkoch78 in #105
• Add llm-filter by @aflament in #106
• Add logs into context for LLM {Evaluator, Controller, Filter, Similarity Scorer} by @aflament in #107
• Release 0.0.14 by @aflament in #108

Full Changelog: v0.0.13...v0.0.14

v0.0.13

What's Changed

• CU-8685u1cq8_Council-add-DoWhile-runner by @Naccomy in #94
• Cu 8685uwxba council add while runner by @Naccomy in #95
• Add ranking to Controller's execution unit, enabling parallel chain execution by @gkoch78 in #96
• Controller Parallelism by @aflament in #97
• Feature release 0.0.13 by @aflament in #98

Full Changelog: v0.0.12...v0.0.13

v0.0.12

What's Changed

• Feature bugfix budget by @gkoch78 in #83
• Add code examples for static graph and execution log by @gkoch78 in #84
• Message Banner on doc built from main by @gkoch78 in #85
• Feature refactor runner tests by @gkoch78 in #86
• Feature fix if runner missing monitor dependency and add an else runner by @gkoch78 in #87
• Bug fix wrong runner node in agent chain graph by @gkoch78 in #88
• Feature add helper monitored llm by @gkoch78 in #90
• Feature exporting type hints by @gkoch78 in #91
• Feature add logging by @gkoch78 in #92
• Support Chains with instructions by @aflament in #89
• Version 0.0.12 by @aflament in #93

Full Changelog: v0.0.11...v0.0.12

v0.0.11

Highlighted New Features

• Export of the agent execution graph as Json
• Introduction of an execution log that keeps track of all execution duration, messages, and budget.
• Consolidation of Context
• Introduction of Filter

Know Issues

• budget consumptions are double counted (fixed by : #83)
• budget consumptions are not tracked in the execution log (fixed by: #83)
• execution log is not publicly available yet

Breaking changes

This release contains a number of breaking changes, including:

• Budget class has moved from council.runners to council.context
• Controller was holding two responsibilities: get_plan and select_responses. It is now only responsible for the planning through the execute function.
• Filter is a new concept which takes on the select_responses responsibility from the Controller through its own execute function
• The budget parameter of all execute functions has been removed. The budget is now available through the context, context.budget
• execute function should no longer be overridden in custom class implementation. Override _execute instead
• Chains are now passed to Controller.__init__ instead of Agent.__init__
• context.chatHistory has been renamed context.chat_history
• context.evaluationHistory has been replaced with context.evaluation
• llm.post_chat_request has changed to accept an LLMContext

What's Changed

• Cu 8685duxrx restructure documentation by @Naccomy in #70
• CU-8685duz8p_Add-Parallel-Execution-example by @Naccomy in #71
• CU-8685duz6t_Add-Token-Budgets-example by @Naccomy in #72
• Feature extract filter from controller by @aflament in #73
• Feature track agent execution information by @gkoch78 in #74
• Dependencies Updates by @aflament in #76
• Feature add notebook test by @gkoch78 in #77
• Update README.md by @aflament in #78
• Add a name property on Agent by @gkoch78 in #79
• Ensure budget is not expired if a remaining is equal to 0 by @aflament in #80
• Update documentation for context objects by @gkoch78 in #81
• Release 0.0.11 by @aflament in #82

New Contributors

• @Naccomy made their first contribution in #70

Full Changelog: v0.0.10...v0.0.11

v0.0.10

What's Changed

• Readme and documentation Updates by @aflament in #60
• Feature improve templating by @aflament in #61
• Fix readme badges by @aflament in #62
• Feature enhance llm evaluation by @aflament in #63
• Feature add llm fallback mechanism by @aflament in #64
• Add unit test for openai model configuration by @gkoch78 in #65
• Feature adapt prompt for llama2 by @aflament in #66

Full Changelog: v0.0.9...v0.0.10