Added

• Added support for using multiple glossaries via the glossary_ids parameter
  (up to 5 glossaries) in translate_text(), translate_document(),
  translate_document_upload(), and translate_document_from_filepath().
• Added support for style rules in document translation via the style_rule
  parameter in translate_document(), translate_document_upload(), and
  translate_document_from_filepath().
• Added support for translation memories in document translation via the
  translation_memory and translation_memory_threshold parameters in
  translate_document(), translate_document_upload(), and
  translate_document_from_filepath().
• Added --glossary-ids CLI argument for the text and document commands,
  allowing up to 5 glossaries to be used per translation.
• Added --style-id, --translation-memory-id, and
  --translation-memory-threshold CLI arguments to the document command.
• Added support for the translation memory management APIs:
  get_translation_memory(), list_translation_memory_segments(),
  delete_translation_memory(), create_translation_memory_import(),
  upload_translation_memory_file(), create_translation_memory_export(),
  get_translation_memory_job(),
  wait_until_translation_memory_job_done(), and
  download_translation_memory_export(), along with the
  import_translation_memory_from_filepath() and
  export_translation_memory_to_filepath() convenience functions that create
  the job, transfer the TMX file, and wait for the job to finish. Because the
  API detects the file upload asynchronously, an import job keeps reporting
  awaiting_input for a while after the upload; the wait loop polls through
  that status, so pass timeout_s to bound the wait.
• Added the TranslationMemorySegments, TranslationMemorySegment,
  TranslationMemoryTargetSegment, TranslationMemoryImport,
  TranslationMemoryExport, TranslationMemoryJob, and
  TranslationMemoryJobResult types.
• Added creation_time and updated_time properties to
  TranslationMemoryInfo.
• Added the translation-memory CLI command with the list, get,
  segments, import, export, job, and delete subcommands.

Added

• Added support for translation memories in text translation via
  translation_memory and translation_memory_threshold parameters in
  translate_text().
• Added list_translation_memories() method to the DeepLClient class for
  listing available translation memories.
• Added --translation-memory-id and --translation-memory-threshold CLI
  arguments.

Fixed

• Fixed "gitlab release" CI job to explicitly set the release-cli image, avoiding issues with missing $HOME from the global image.

Added

• Added new language constants from January 2026 API release of 81 new languages.
• Added support for style rules CRUD endpoints in the DeepLClient class:
  create_style_rule(), get_style_rule(), update_style_rule_name(),
  update_style_rule_configured_rules(), and delete_style_rule().
• Added support for style rule custom instruction CRUD endpoints in the
  DeepLClient class: create_style_rule_custom_instruction(),
  get_style_rule_custom_instruction(),
  update_style_rule_custom_instruction(), and
  delete_style_rule_custom_instruction().
  Please refer to the README for usage instructions.

Changed

• Updated formality tests to accept either formal or informal output when using default formality,
  since the default formality is automatic.

Changed

• Updated three tests to be less flakey and reflect new behavior regarding the model type.
• Improved NotFoundException error message by removing the misleading "check server_url" suggestion.

Fixed

• Fixed a bug where translating with style rules would sometimes incorrectly result in
  a "target_lang must match style rule language" error

Security

• Updated dependencies to fix CVE-2025-66418 and CVE-2025-66471

Added

• Added tag_handling_version parameter to translate_text() to specify which version
  of the tag handling algorithm to use. Options are v1 and v2.
• Added an example CLI script for realtime audio translation using DeepL's Voice API.

Added

• Added custom_instructions parameter to translate_text() to customize translation behavior with up to 10 instructions (max 300 characters each). Only supported for target languages: de, en, es, fr, it, ja, ko, zh and their variants. Note: using the custom_instructions parameter will use the quality_optimized model type as the default. Requests combining custom_instructions and the latency_optimized model type will be rejected.

• Added support for the GET /v3/style_rules endpoint in the client library, the implementation can be found in the DeepLClient class. Please refer to the README for usage instructions
• Added style_id option to translate_text() which allows text translation with style rules.

Changed

• extra_body_parameters will now overwrite already set HTTP parameters

Added

• Added feature to add arbitrary body parameters in translate_text

Deprecated

• Dropped support for Python 3.8 and older.

Security

• Updated requests minimum version to 2.32.4 to resolve security advisories
  CVE-2024-47081 and CVE-2024-35195.

Added

• Added support for the /v3 Glossary APIs in the client library while providing backwards compatibility for the previous /v2 Glossary endpoints. Please refer to the README for usage instructions.