open-llm-leaderboard/open_llm_leaderboard

Resource: Understanding the new benchmarks

pinned

1

#796 opened about 1 month ago by

rombodawg

💬 Discussion thread: Model contamination techniques 💬

pinned

34

#472 opened 8 months ago by

clefourrier

💎 Resources and community initiatives around the Leaderboard! 💎

pinned

#174 opened 12 months ago by

clefourrier

re-evaluating failed models after fixing the NaN weights

#869 opened about 1 hour ago by

awnr

Parameter count at 0 for incorrectly detected params is misleading

5

#867 opened 1 day ago by

dnhkng

What should a finetuned model's license be if the model is MIT but the datasets are Apache 2.0 and cc-by-4.0

3

#866 opened 1 day ago by

rasyosef

Increasing upper limit of `Select the number of parameters (B)` to support larger open-source models like `meta-llama/Meta-Llama-3.1-405B-Instruct`

5

#858 opened 9 days ago by

singhsidhukuldeep

Model deleted from Pending

16

#850 opened 15 days ago by

dnhkng

Upvote to evaluate deepseek-coder-v2

2

#793 opened about 1 month ago by

g1y5x3

Feature request: Add toggle to only show models with linked dataset

1

#763 opened 2 months ago by

ThiloteE

Feature request: Hide models with insufficient model card from default view in leaderboard

4

#762 opened 2 months ago by

ThiloteE

Discussion: naming pattern to converge on to better identify fine-tunes

17

#761 opened 2 months ago by

ThiloteE

Crowd-Source Hardware for the LeaderBoard?

4

#570 opened 6 months ago by

ibivibiv

Feature request: Using weights hash to identify duplicates

1

#422 opened 8 months ago by

mrfakename

Tool: Adding evaluation results to model cards

47

#370 opened 9 months ago by

Weyaxi

Feature suggestion: average of selected (rather than all) columns

4

#368 opened 9 months ago by

Minus0

Tool: Open LLM Leaderboard Model Renamer

31

#310 opened 10 months ago by

Weyaxi