Leaderboard Update, in sync with BFCL April 27th Release #391
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
CharlieJCJ
approved these changes
Apr 27, 2024
ShishirPatil
pushed a commit
that referenced
this pull request
Apr 27, 2024
In this PR, we fix some inconsistency issues in the cost and latency calculation for open-source models, which are now all calculated when serving the model with [vLLM](https://github.com/vllm-project/vllm) using 8 V100 GPUs. $$\text{Cost} = \text{Latency per 1000 function call} * (\text{8xV100 azure-pay-as-you-go-price per hour / 3600})$$ This PR **DOES** change the leaderboard value in the `cost` and `latency` columns; but it **DOES NOT** change the accuracy score. We will update the leaderboard in a different PR #391. We want to thank the community for pointing out this oversight. Thanks [@abacaj](https://twitter.com/abacaj) and [@teknium1](https://twitter.com/Teknium1) for initially raising the issue, and thanks [@natikgadzhi](https://twitter.com/natikgadzhi) [@HamelHusain](https://twitter.com/HamelHusain) [@nicoritschel](https://twitter.com/nicoritschel) [@winglian](https://twitter.com/winglian) [@olafgeibig](https://twitter.com/olafgeibig) and many others for joining the conversation. We are listening to community feedback and continuously improving our Berkeley Function Calling Leaderboard. Discussions like [this](https://twitter.com/abacaj/status/1784003306508980250) serve as great examples. Let us know what you want us to include next! --------- Co-authored-by: Charlie Cheng-Jie Ji <[email protected]> Co-authored-by: Fanjia Yan <[email protected]>
devanshamin
pushed a commit
to devanshamin/gorilla
that referenced
this pull request
Jul 9, 2024
…rPatil#390) In this PR, we fix some inconsistency issues in the cost and latency calculation for open-source models, which are now all calculated when serving the model with [vLLM](https://github.com/vllm-project/vllm) using 8 V100 GPUs. $$\text{Cost} = \text{Latency per 1000 function call} * (\text{8xV100 azure-pay-as-you-go-price per hour / 3600})$$ This PR **DOES** change the leaderboard value in the `cost` and `latency` columns; but it **DOES NOT** change the accuracy score. We will update the leaderboard in a different PR ShishirPatil#391. We want to thank the community for pointing out this oversight. Thanks [@abacaj](https://twitter.com/abacaj) and [@teknium1](https://twitter.com/Teknium1) for initially raising the issue, and thanks [@natikgadzhi](https://twitter.com/natikgadzhi) [@HamelHusain](https://twitter.com/HamelHusain) [@nicoritschel](https://twitter.com/nicoritschel) [@winglian](https://twitter.com/winglian) [@olafgeibig](https://twitter.com/olafgeibig) and many others for joining the conversation. We are listening to community feedback and continuously improving our Berkeley Function Calling Leaderboard. Discussions like [this](https://twitter.com/abacaj/status/1784003306508980250) serve as great examples. Let us know what you want us to include next! --------- Co-authored-by: Charlie Cheng-Jie Ji <[email protected]> Co-authored-by: Fanjia Yan <[email protected]>
aw632
pushed a commit
to vinaybagade/gorilla
that referenced
this pull request
Aug 22, 2024
…rPatil#390) In this PR, we fix some inconsistency issues in the cost and latency calculation for open-source models, which are now all calculated when serving the model with [vLLM](https://github.com/vllm-project/vllm) using 8 V100 GPUs. $$\text{Cost} = \text{Latency per 1000 function call} * (\text{8xV100 azure-pay-as-you-go-price per hour / 3600})$$ This PR **DOES** change the leaderboard value in the `cost` and `latency` columns; but it **DOES NOT** change the accuracy score. We will update the leaderboard in a different PR ShishirPatil#391. We want to thank the community for pointing out this oversight. Thanks [@abacaj](https://twitter.com/abacaj) and [@teknium1](https://twitter.com/Teknium1) for initially raising the issue, and thanks [@natikgadzhi](https://twitter.com/natikgadzhi) [@HamelHusain](https://twitter.com/HamelHusain) [@nicoritschel](https://twitter.com/nicoritschel) [@winglian](https://twitter.com/winglian) [@olafgeibig](https://twitter.com/olafgeibig) and many others for joining the conversation. We are listening to community feedback and continuously improving our Berkeley Function Calling Leaderboard. Discussions like [this](https://twitter.com/abacaj/status/1784003306508980250) serve as great examples. Let us know what you want us to include next! --------- Co-authored-by: Charlie Cheng-Jie Ji <[email protected]> Co-authored-by: Fanjia Yan <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
As mentioned in #390, in this PR, we fix some inconsistency issues in the cost and latency calculation for open-source models, which are now all calculated when serving the model with vLLM using 8 V100 GPUs.$$\text{Cost} = \text{Latency per 1000 function call} * (\text{8xV100 azure-pay-as-you-go-price per hour / 3600})$$
We want to thank the community for pointing out this oversight. Thanks @abacaj and @Teknium1 for initially raising the issue, and thanks @natikgadzhi @HamelHusain @nicoritschel @winglian @olafgeibig and many others for joining the conversation. We are listening to community feedback and continuously improving our Berkeley Function Calling Leaderboard. Discussions like this serve as great examples. Let us know what you want us to include next!
This PR DOES change the leaderboard scores for
costsandlatency, but notaccuracy.Co-authored-by: Charlie Cheng-Jie Ji [email protected]
Co-authored-by: Fanjia Yan [email protected]