Flexeval

Latest version: v0.8.1

Safety actively analyzes 693883 Python packages for vulnerabilities to keep your Python projects secure.

Scan your dependencies

Page 1 of 6

0.8.1

What's Changed
* Upgrade vLLM from v0.5 series to v0.6 by butsugiri in https://github.com/sbintuitions/flexeval/pull/103
* Make some modules accessible from the top level by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/104
* Debug `SequenceClassificationRewardModel` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/105


**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.8.0...v0.8.1

0.8.0

What's Changed
* Update reward dataset with `TemplateRewardBenchDataset` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/96
* Add default values for instance classes by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/97
* Set `require_incremental_response` to `False` by default by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/98
* Fix the default value of template based reward datasets by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/100
* Update `RewardBenchInstance` to handle a list of messages by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/99
* Implement `SequenceClassificationRewardModel` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/101
* Implement RewardModel based on log probs by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/102
* End of life of Python 3.8 by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/80


**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.7.8...v0.8.0

0.7.8

What's Changed
* EvalSetupの更新 by teruaki-o in https://github.com/sbintuitions/flexeval/pull/92
* hellaswagの few-shot examples を修正 by Ktakuya332C in https://github.com/sbintuitions/flexeval/pull/93
* PIQAを追加 by Ktakuya332C in https://github.com/sbintuitions/flexeval/pull/94
* ARC Challenge を追加 by Ktakuya332C in https://github.com/sbintuitions/flexeval/pull/95

New Contributors
* teruaki-o made their first contribution in https://github.com/sbintuitions/flexeval/pull/92

**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.7.7...v0.7.8

0.7.7

What's Changed
* Modify `OpenAIChatBatchAPI` to logging Batch IDs. by junya-takayama in https://github.com/sbintuitions/flexeval/pull/90
* Fix bug: `system_message` rendering issue in `ChatLLMScore` and `ChatLLMLabel` by junya-takayama in https://github.com/sbintuitions/flexeval/pull/91


**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.7.6...v0.7.7

0.7.6

What's Changed
* Classification-based evaluation using `LanguageModel` by junya-takayama in https://github.com/sbintuitions/flexeval/pull/88
* Add evaluate_module option to CodeEval class by Ktakuya332C in https://github.com/sbintuitions/flexeval/pull/89


**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.7.5...v0.7.6

0.7.5

What's Changed
* Update openai version to 1.52.2 or higher. by junya-takayama in https://github.com/sbintuitions/flexeval/pull/87


**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.7.4...v0.7.5

Page 1 of 6

© 2025 Safety CLI Cybersecurity Inc. All Rights Reserved.