v0.4.0-python

• Fix a bug that makes Engine's copy and deepcopy method returning InternalEngine rather than Engine.

v0.3.13-python

• Changes from v0.5.5-rust

v0.5.5-rust

What's Changed

• Support complement of a regular expression in KBNF.
• Make warning less prominent for duplicate tokens. by @andrejonasson in #20

Full Changelog: v0.5.4-rust...v0.5.5-rust

v0.3.12-python

• Add check_if_token_allowed in Python FFI to facilitate workarounds in formatron that handles certain weird behaviors in integrations.

v0.3.11-python

• Fix issue on CPU in mask_logits_fast. Only pin if cuda device type. by @andrejonasson in #19
• Make cached tensors in mask_logits_fast actually pinned on cuda.

New Contributors

• @andrejonasson made their first contribution in #19

Full Changelog: v0.3.10-python...v0.3.11-python

v0.3.10-python

• Significantly improved torch tensor masking speed

v0.3.9-python

• Bug fix: incorrect logits mask will no longer be produced when multiple different language models' vocabulary with the same logits length are used to create engines.

v0.5.3-rust

• Bug fix: #substrs now correctly constrain the allowed tokens

v0.5.2-rust

• Bug fix: undefined start nonterminal in kbnf grammar string will no longer panick. Instead, an error will be returned.

v0.3.8-python

• Changes from v0.5.3-rust