News 1 : About delay in Finished evaluation

#3
by alielfilali01 - opened
Open Arabic LLM Leaderboard org

We've encountered a slight hiccup with our model evaluations, due to memory constraints, our evaluations were temporarily halted, but rest assured, we're on top of it. However, we’ve expanded our capacity with 7 new instances and relaunched the evaluations yesterday.

Currently, the evaluations for the Jais (13b chat, 30b chat v1, 30b chat v3) models, 3 Qwen models, and Falcon 40 are underway. These models are quite large, so please expect results to come in gradually.

Also, we are closely monitoring the process every hour to ensure everything runs smoothly.

Stay tuned for more updates!

alielfilali01 changed discussion status to closed

Sign up or log in to comment