Releases: aurelio-labs/semantic-router
Release v0.0.21
Release v0.0.21
Key new features include MistralAI API for Mistral LLMs and encoders by @dwmorris11, support for more scalable and persistent route layers using Pinecone by @Siraj-Aizlewood and @jamescalam, and new splitting mechanisms for Conversation objects to improve context support for routing decisions by @Siraj-Aizlewood.
What's Changed
- feat: Added Method for Deletion of Routes in RouteLayer by @Siraj-Aizlewood in #128
- feat: Incorporating Semantic Topic Splitters into Conversation by @Siraj-Aizlewood in #134
- fix: pip install error in 03-basic-langchain-agent.ipynb by @jzcruiser in #131
- fix: issue #94 added from clause to the exception statement in by @dwmorris11 in #143
- feat: Added support for MistralAI API. This includes a by @dwmorris11 in #142
- feat: separate indexes and PineconeIndex by @jamescalam in #146
- fix: Saving JSON/YAML of Layer Config by @Siraj-Aizlewood in #145
New Contributors
- @jzcruiser made their first contribution in #131
- @dwmorris11 made their first contribution in #143
Full Changelog: v0.0.20...v0.0.21
Contributors
Assets 2
Release v0.0.20
Contributors
Assets 2
Release v0.0.19
What's New
We've added more flexible and accurate routes. By using route-specific score thresholds and the new route layer fit and evaluate methods, we can get optimal route layer performance in seconds! We have also added a few new example notebooks and support for OpenAI's 3rd generation embedding models and the new dimensions parameter that they introduce.
What's Changed
- feat: Optimal Route-Level Score Thresholds by @Siraj-Aizlewood in #121
- feat: new examples and openai upgrade by @jamescalam in #126
Full Changelog: v0.0.18...v0.0.19
Contributors
Assets 2
v0.0.18
We now have Pydantic v2 support, this should help those of you using LangChain 0.1.x. For enterprise users, thanks to @arashaga we can now run both the encoder and LLM for dynamic routes using just Azure OpenAI!
What's Changed
- feat: Allow Pydantic V2+ by @kdcokenny in #109
- chore: update README.md and add course by @jamescalam in #119
- feat: Added AzureOpenAILLM by @arashaga in #112
- feat: Return triggered flag in response by @ashraq1455 in #123
- chore: James/0018 by @jamescalam in #124
New Contributors
Full Changelog: v0.0.17...v0.0.18
Contributors
Assets 2
v0.0.17
Primarily bug fix for llama.cpp ModuleNotFoundError
What's Changed
- fix: Fixes ModuleNotFoundError by @bruvduroiu in #110
- fix: issue #94 raise from exceptions by @alongadot in #105
- feat: Update version to 0.0.17 by @simjak in #113
- chore: Update for version by @jamescalam in #114
New Contributors
- @alongadot made their first contribution in #105
Full Changelog: v0.0.16...v0.0.17
Contributors
Assets 2
Release v0.0.16
As part of the latest updates to the Semantic Router, we've made significant strides in enhancing its capabilities and compatibility:
-
Full Local Support: We've added support for Local LLMs via LlamaCpp so that we can run dynamic routes fully local. Our initial tests show Mistral 7B with our grammar implementation to outperform GPT-3.5 as a dynamic route decision maker.
-
New Encoder Integrations: We're excited to announce the addition of the HuggingFace, TFIDF, and Azure OpenAI Encoders. These integrations significantly expand the router's ability to interface with a variety of language models.
-
Enhanced Python Compatibility: In this release, we've ensured backward-compatibility with Python 3.9, we expect to add full compatibility for Python 3.12 soon.
These developments reflect our ongoing effort to enhance the Semantic Router's language processing abilities and its adaptability across various platforms. We're proud to bring these features to our community and are grateful for the contributions from new and existing contributors.
New Contributors
- @digriffiths made their first contribution in #38
- @mckeown12 made their first contribution in #73
What's Changed
- fix: minor tweak to 02 doc by @jamescalam in #87
- feat: Add HuggingFace Encoder by @ashraq1455 in #90
- feat: Compatible with Python 3.9 by @simjak in #92
- fix: improve hard coded prompts by @hananell in #100
- feat: Add tfidf by @digriffiths in #38
- chore: Typings in uppecase by @hananell in #101
- feat: Add AzureOpenAIEncoder by @mckeown12 in #73
- feat: Adds LlamaCpp LLM by @bruvduroiu in #96
- chore: 16 release by @jamescalam in #106
Full Changelog: v0.0.15...v0.0.16
Contributors
Assets 2
Release v0.0.15
What's Changed
This release is focused on compatibility with popular services while enabling more of the library to be run locally — we are keeping the library lightweight with dependency groups. We have begun implementing classes for Conversation and Message objects, these integrate with the new semantic text splitter — which is a key component in enabling chat history for semantic router.
⭐️ PR Highlights
- feat: Add Semantic Splitter by @ashraq1455 in #72
- feat: Add fastembed encoder by @kdcokenny in #68
- feat: Add OpenAI, Cohere and OpenRouter as separate LLMs by @ashraq1455 in #79
👏🏼 New Contributors
- @avatsaev made their first contribution in #54
- @kdcokenny made their first contribution in #68
- @ashraq1455 made their first contribution in #72
- @hananell made their first contribution in #84
PRs
- Update 00-introduction.ipynb by @jamescalam in #45
- Update 01-save-load-from-file.ipynb by @jamescalam in #46
- Update 02-dynamic-routes.ipynb by @jamescalam in #47
- Update 03-basic-langchain-agent.ipynb by @jamescalam in #48
- Update README.md by @jamescalam in #49
- Update README to show latest syntax by @jamescalam in #51
- Fix for failing black lint in docs by @jamescalam in #50
- Adding link to docs by @jamescalam in #52
- Update 00-introduction.ipynb by @avatsaev in #54
- Adding Python versions badge by @jamescalam in #55
- enforcing conventional commits by @simjak in #57
- fix: remove unnecessary files by @simjak in #58
- feat: Add fastembed encoder by @kdcokenny in #68
- feat: Add Semantic Splitter by @ashraq1455 in #72
- fix: Coventional commits for PR titles by @simjak in #74
- feat: Updated function calling example by @simjak in #75
- feat: created fastembed and hybrid dependency groups by @jamescalam in #81
- feat: Add OpenAI, Cohere and OpenRouter as separate LLMs by @ashraq1455 in #79
- fix: Gitignore jetbrains by @hananell in #84
- feat: Move score_threshold to encoders by @ashraq1455 in #82
- fix: James/v0.0.15 by @jamescalam in #85
Full Changelog: v0.0.14...v0.0.15
Contributors
Assets 2
Release v0.0.14
Contributors
Assets 2
Release v0.0.11
Merge pull request #32 from aurelio-labs/simonas/assistant OSS Assistant example
Release v0.0.10
Merge pull request #33 from aurelio-labs/luca/fix-on-embeddings-check Fix for embeddings