This repository was archived by the owner on Jul 4, 2025. It is now read-only.

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.

...

1 commit
2 files changed
1 contributor

Commits on Jun 12, 2025

Hostfix: remove not needed params from load_model (#2209 )

* refactor: remove --pooling flag from model loading

The --pooling flag was removed as the mean pooling functionality not needed in chat models. This fixes the regression

* feat(local-engine): add ctx_len parameter support

Adds support for the ctx_len parameter by appending --ctx-size with its value. Removed outdated parameter mappings from the kParamsMap to reflect current implementation details and ensure consistency.

* feat: add conditional model parameters based on path

When the model path contains both "jan" and "nano" (case-insensitive), automatically add
speculative decoding parameters to adjust generation behavior. This improves
flexibility by enabling environment-specific configurations without manual
parameter tuning. Also includes necessary headers for string manipulation and
fixes whitespace in ctx_len handling.

* chore: remove redundant comment

The comment was redundant as the code's purpose is clear without it, improving readability.

qnixsynapse authored Jun 12, 2025