Repository Issues
huggingface/lighteval
Lighteval is your all-in-one toolkit for evaluating LLMs across multiple backends
Issues
Open
[EVAL] Long Horizon Execution
good first issuehelp wantednew-task
Has a beginner-friendly label
7 comments0 reactions1 assignee
Open
[EVAL] Add kyrgyzLLM benchmark
good first issuenew-task
Why recommendedHas a beginner-friendly label
Has a beginner-friendly label
1 comment0 reactions1 assignee
Open
[FT] showing count in Markdown summary table
featuregood first issue
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
5 comments1 reaction0 assignees
Open
[FT] Add tests for nanotron
featuregood first issuehelp wantedignore-for-release
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
3 comments0 reactions0 assignees
Open
[BUG] custom model docs don't run: missing imports
bugdocumentationgood first issue
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
5 comments0 reactions0 assignees
Open
[FT] Manage script and language in the Language enum
featuregood first issue
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
2 comments0 reactions0 assignees
Open
[EVAL] TauBench:
help wantednew-task
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
1 comment0 reactions0 assignees
Open
[EVAL] SciCode: reasearch coding benchmark
help wantednew-task
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
2 comments0 reactions0 assignees
Open
[BUG] Optimize tokenization
buggood first issuehelp wantedscience-team
Why recommendedHas a beginner-friendly label
Has a beginner-friendly label
1 comment0 reactions1 assignee
Open
[EVAL] HELMET: long context evals
help wantednew-task
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
2 comments0 reactions0 assignees
Open
[EVAL] SWEBENCH multilingual
help wantednew-task
Why recommendedNo assignee yet · No comments yet
No assignee yetNo comments yetHas a beginner-friendly label
0 comments0 reactions0 assignees
Open
[FT] Add tests for `VLLMModel` base methods
featuregood first issue
Why recommendedNo comments yet · Has a beginner-friendly label
No comments yetHas a beginner-friendly label
0 comments1 reaction1 assignee
Open
Call for contributions: Translate lighteval's doc into Chinese
good first issuehelp wantednew-task
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
2 comments1 reaction0 assignees
Open
[BUG] Installing lighteval breaks hydra-core
bughelp wanted
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
1 comment1 reaction0 assignees
Open
[EVAL] Adding PHARE
good first issuehelp wantednew-task
Why recommendedHas a beginner-friendly label
Has a beginner-friendly label
9 comments0 reactions1 assignee
Open
[FT] Improve Documentation and Examples
documentationfeaturegood first issuehelp wanted
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
6 comments1 reaction0 assignees
Open
[FT] Log progress bar on main process
featurehelp wantedscience-team
Why recommendedNo assignee yet · No comments yet
No assignee yetNo comments yetHas a beginner-friendly label
0 comments0 reactions0 assignees
Open
[FT] Build in a way to specify specific IDs/Lines in Dataset to use as few-shot examples in the same split
featuregood first issuehelp wanted
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
3 comments0 reactions0 assignees
Open
[EVAL] Big-Bench Extra Hard (BBEH)
good first issuehelp wantednew-taskscience-team
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
3 comments0 reactions0 assignees
Open
[EVAL] Add TUMLU benchmark
good first issuehelp wantednew-task
Why recommendedNo assignee yet · Has a beginner-friendly label
No assignee yetHas a beginner-friendly label
10 comments0 reactions0 assignees