opencompass
opencompass copied to clipboard
OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.
### Prerequisite - [X] I have searched [Issues](https://github.com/open-compass/opencompass/issues/) and [Discussions](https://github.com/open-compass/opencompass/discussions) but cannot get the expected help. - [X] The bug has not been fixed in the [latest version](https://github.com/open-compass/opencompass). ### Type...
### Describe the feature METRIC_WHITELIST怎么添加白名单 ### Will you implement it? - [ ] I would like to implement this feature and create a PR!
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand...
### Describe the feature When using a Hugging Face model with the parameter num_return_sequences set greater than 1, the output column “predictions” becomes a list instead of a string. As...
### 描述该功能 is there any way to evaluate an agent? btw how to use transbench? ### 是否希望自己实现该功能? - [ ] 我希望自己来实现这一功能,并向 OpenCompass 贡献代码!
### Describe the feature # Motivation Currently, PyTorch is the only supported framework for OpenCompass. We would like to propose supporting different frameworks such as MindSpore on Ascend NPU in...
### Prerequisite - [x] I have searched [Issues](https://github.com/open-compass/opencompass/issues/) and [Discussions](https://github.com/open-compass/opencompass/discussions) but cannot get the expected help. - [x] The bug has not been fixed in the [latest version](https://github.com/open-compass/opencompass). ### Type...
### 描述该功能 The model evaluation is too slow. Can we turn off the model thinking to improve the evaluation speed? Is this option available for configuration ### 是否希望自己实现该功能? - [...
### Prerequisite - [x] I have searched [Issues](https://github.com/open-compass/opencompass/issues/) and [Discussions](https://github.com/open-compass/opencompass/discussions) but cannot get the expected help. - [x] The bug has not been fixed in the [latest version](https://github.com/open-compass/opencompass). ### Type...