Name | Modified | Size | Downloads / Week |
---|---|---|---|
Parent folder | |||
README.md | 2025-07-12 | 2.3 kB | |
release 0.3.4 source code.tar.gz | 2025-07-12 | 11.1 MB | |
release 0.3.4 source code.zip | 2025-07-12 | 11.2 MB | |
Totals: 3 Items | 22.3 MB | 0 |
What's Changed
Breaking changes
arch_config file format change
In llm_providers section we now allow model to contain provider as part of the model definition. This is to simplify the llm_providers section and to allow more concise way to defining providers,
Here is a sample llm_provider definition after this change,
- access_key: $OPENAI_API_KEY
model: openai/gpt-4o-mini
- access_key: $OPENAI_API_KEY
model: openai/gpt-4o
default: true
Before 0.3.4 you'd have to do following,
- name: gpt-4o-mini
access_key: $OPENAI_API_KEY
provider_interface: openai
model: gpt-4o-mini
- name: gpt-4o
access_key: $OPENAI_API_KEY
provider_interface: openai
model: gpt-4o
default: true
preference based routing
You can now specify preference based routing with route names as part of llm_providers (notice name/description),
- model: openai/gpt-4o
access_key: $OPENAI_API_KEY
routing_preferences:
- name: code understanding
description: understand and explain existing code snippets, functions, or libraries
- model: openai/gpt-4.1
access_key: $OPENAI_API_KEY
routing_preferences:
- name: code generation
description: generating new code snippets, functions, or boilerplate based on user prompts or requirements
Other changes
- Salmanap/chrome extension routing by @salmanap in https://github.com/katanemo/archgw/pull/511
- added the tabs permission again by @salmanap in https://github.com/katanemo/archgw/pull/521
- removing the tabs permission and reverting to window.postMessage by @salmanap in https://github.com/katanemo/archgw/pull/524
- better model names by @adilhafeez in https://github.com/katanemo/archgw/pull/517
- release 0.3.4 by @adilhafeez in https://github.com/katanemo/archgw/pull/525
Full Changelog: https://github.com/katanemo/archgw/compare/0.3.3...0.3.4